Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development and application of a non-rejectable islet graft using indoleamine 2,3-dioxygenase (IDO) in… Hosseinitabatabaei, Azadeh 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_spring_hosseinitabatabaei_azadeh.pdf [ 25.56MB ]
Metadata
JSON: 24-1.0167393.json
JSON-LD: 24-1.0167393-ld.json
RDF/XML (Pretty): 24-1.0167393-rdf.xml
RDF/JSON: 24-1.0167393-rdf.json
Turtle: 24-1.0167393-turtle.txt
N-Triples: 24-1.0167393-rdf-ntriples.txt
Original Record: 24-1.0167393-source.json
Full Text
24-1.0167393-fulltext.txt
Citation
24-1.0167393.ris

Full Text

 DEVELOPMENT AND APPLICATION OF A NON-REJECTABLE ISLET GRAFT USING INDOLEAMINE 2,3-DIOXYGENASE (IDO) IN DIABETES.  by  AZADEH HOSSEINITABATABAEI Pharm.D., Tehran University of Medical Sciences, 2008   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIRMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND  POSTDOCTORAL STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2014   © Azadeh Hosseinitabatabaei, 2014 ! ""!ABSTRACT  Type 1 diabetes (T1D) is a devastating disease affecting more than 22 million people worldwide. Early diagnosis and treatment of T1D requires a deep understanding of the mechanisms underlying the progression of autoimmune diabetes. It has been shown that impaired IFN!-induced indoleamine 2,3-dioxygenase (IDO) expression in dendritic cells of early prediabetic female nonobese diabetic (NOD) mice could contribute to their defective self-tolerance. IDO is an immunomodulatory and rate-limiting enzyme in tryptophan catabolism. Since fibroblasts are important components of islet extracellular matrix (ECM), we investigated IFN!-induced IDO expression in NOD dermal fibroblasts. Our findings indicate that IFN! fails to induce IDO expression in NOD dermal fibroblasts and moreover that the mechanism underlying this defect involves defective STAT1 phosphorylation in the IFN!-induced-IDO signaling pathway. We further showed that an IFN!-independent IDO expression pathway is operative in NOD mice fibroblasts.  Islet transplantation, a promising strategy to restore efficient insulin regulation in T1D, is limited by poor post-transplant islet survival and toxicity of immunosuppressants. We developed a novel bioengineered cross-linked collagen matrix (CCM) to act as an ECM for the islets and found that islet function and survival significantly improved with this scaffold.  We previously showed that in an IDO-mediated microenvironment, infiltrated immune cells but not islet cells cannot survive or proliferate. Here, we used IDO to generate a local immunomodulated environment, in which transplanted islets could remain viable and protected without compromising systemic immunity. We developed an improved islet allograft composed of stable IDO-expressing dermal fibroblasts (by lentiviral transduction) ! """!and allogeneic islets embedded within CCM in streptozotocin-induced diabetic mice. This IDO-expressing matrix did not compromise islet function or survival. IDO expression significantly suppressed the proliferation of alloantigen-stimulated splenocytes. Finally, we showed that local IDO expression delivered by a lentiviral vector significantly prolonged islet allograft survival by increasing the population of FOXP3+ cells at the graft site and preventing T cell infiltration.  Overall, the studies in this thesis showed how defects in the IDO signaling pathway may underlie autoimmune diabetes and highlighted the therapeutic potential of IDO expression and improved matrices for enhancing survival and function of islet transplants in T1D.              ! "#!PREFACE  Animal studies were reviewed and approved by the University of British Columbia Committee on Animal Care (Protocol# A10-0372). A. Hosseini-Tabatabaei was trained for Animal Ethics online course (#3266-09) and practical course of Rodent Biology and Husbandry (#RBH-1095-10). The human IDO gene (NM_002164) was a generous gift from Dr. JM Carlin of Miami University. Dr Y. Li generated the Ad-IDO vector and Dr R.T. Kilani constructed the Lenti-IDO vector.  A part of Chapter 1 has been published as a review article in Hosseini-Tabatabaei A. et al.,  (2011) Immunomodulatory role of IDO in physiological and pathological conditions, Current Trends in Immunology. 11:51-63. In Chapter 3, A. Hosseini-Tabatabaei, Dr. R.B. Jalili and Dr. A. Ghahary generated the hypothesis. A. Hosseini-Tabatabaei conceived, designed and performed the experiments, and analyzed the data. Dr. R.B. Jalili contributed to study design, and conduct a part of experiments. Dr. Y. Li contributed to a part of experiments and troubleshooting. Dr. A.M. Rezakhanlou designed specific MHC primers for PCR analysis: Dr. R.T. Kilani prepared NOD dermal fibroblasts. A Hosseini-Tabatabaei wrote and Dr. A. Ghahary edited the manuscript and contributed to the study design. A version of this chapter was published as:  Hosseini-Tabatabaei A., Jalili R.B., Li Y., Kilani R.T., Moeen Rezakhanlou A., Ghahary A. (2012) Mechanism underlying defective interferon gamma-induced IDO expression in non-obese diabetic mouse fibroblasts. PLoS ONE. 7(5):e37747 ! #! In Chapter 4, A. Hosseini-Tabatabaei and Dr. R.B. Jalili conceived, designed and performed the experiments and wrote the manuscript. R. Hartwell designed the original cross-linked collagen matrix. A. Hosseini-Tabatabaei optimized and improved the matrix for better gelation properties. S. Salimi contributed to kynurenine analysis. Dr. A. Ghahary contributed to the study design and edited the manuscript. A version of this chapter was published as: Hosseini-Tabatabaei A., Jalili R.B., Hartwell R., Salimi S., Kilani R.T. Ghahary A. (2013) Embedding islet in a liquid scaffold increases islet viability and function. Canadian Journal of Diabetes, 37(1):27-35 In Chapter 5, A. Hosseini-Tabatabaei conceived and designed and performed the experiments, analyzed the data and wrote the manuscript. Dr. R.B. Jalili contributed designing and performing a part of experiments. Dr. A. Ghahary contributed to the study design and edited the manuscript. Dr. R.T Kilani prepared the lentiviral vector transduced fibroblasts. Dr. Y. Zhang assisted with immunohistochemistry staining. A version of this chapter has been submitted to a peer-reviewed journal as:  Hosseini-Tabatabaei A., Jalili R.B, Hartwell R., Zhang Y., Kilani R.T. Ghahary A. (2013) Immunoprotection and functional improvement of allogeneic islets in diabetic mice, using a stable IDO-producing scaffold.     ! "#!TABLE OF CONTENTS !ABSTRACT ....................................................................................................................... ii PREFACE ......................................................................................................................... iv TABLE OF CONTENTS ................................................................................................ vi LIST OF TABLES ........................................................................................................... xi LIST OF FIGURES ........................................................................................................ xii LIST OF ABBREVIATIONS ....................................................................................... xiv ACKNOWLEDGEMENTS ......................................................................................... xvii DEDICATION ............................................................................................................. xviii !!Chapter 1: INTRODUCTION .................................................................................. 1 ! 1.1 TYPE 1 DIABETES ................................................................................................ 2 1.1.1 Diabetes mellitus ................................................................................................ 2 1.1.2 Pathogenesis of type 1 diabetes ......................................................................... 3 1.1.2.1 Destruction of pancreatic " cells ..................................................................... 5 1.1.2.2 Dysregulation of immunotolerance ................................................................. 6 1.1.2.3 Environmental factors ..................................................................................... 7 1.1.2.4 Genetic factors ................................................................................................ 9 1.1.3 Current treatments .............................................................................................. 9 1.1.4 Animal models of insulin-dependent diabetes mellitus ................................... 11 1.2 ISLET TRANSPLANTATION ........................................................................... 13 1.2.1 Islet isolation .................................................................................................... 13 1.2.2 Current status of islet transplantation .............................................................. 14 1.2.3 Transplantation immunology ........................................................................... 16 1.2.3.1 Antigen recognition pathways ...................................................................... 16 1.2.3.2 Hyperacute, acute and chronic rejections ..................................................... 18 ! "##!1.2.3.3 Role of antibodies, leukocytes and cytokines ............................................... 19 1.2.4 Limitations of islet transplantation .................................................................. 20 1.2.5 Factors affecting transplantation outcome ....................................................... 21 1.2.5.1 Donor variables ............................................................................................. 21 1.2.5.2 Disruption of the islet extracellular matrix ................................................... 22 1.2.5.3 Hypoxia ......................................................................................................... 22 1.2.5.4 Innervation .................................................................................................... 23 1.2.5.5 Instant blood-mediated inflammatory reaction ............................................. 24 1.2.5.6 Glucotoxicity ................................................................................................. 24 1.2.5.7 Endoplasmic reticulum stress ....................................................................... 25 1.2.5.8 Islet amyloid formation ................................................................................. 27 1.2.5.9 Immunosuppressive drugs ............................................................................ 27 1.2.5.10 Immune rejection ........................................................................................ 28 1.2.5.11 Transplantation site ..................................................................................... 28 1.2.6 Current strategies to improve islet transplant outcome .................................... 29 1.2.6.1 Addressing the shortage of islet donors ........................................................ 29 1.2.6.2 Improving islet engraftment and survival ..................................................... 31 1.2.6.3 Improving immunosuppressive strategies ..................................................... 34 1.3 INDOLEAMINE 2,3-DIOXYGENASE .............................................................. 36 1.3.1. Tryptophan breakdown and the role of IDO ................................................... 36 1.3.2 IDO expression ................................................................................................ 38 1.3.3. IDO structure, biochemical characteristics and regulation of expression ....... 39 1.3.4. IDO and modulation of immune response ...................................................... 40 1.3.4.1 IDO and T cells ............................................................................................. 40 1.3.4.2. IDO and antigen presenting cells ................................................................. 41 1.3.4.3. IDO and T regulatory cells .......................................................................... 42 1.3.4.4. IDO and B cells ............................................................................................ 43 1.3.4.5. IDO and NK cells ......................................................................................... 44 1.3.5. Molecular mechanism underlying immunomodulatory effects of IDO .......... 45 1.3.6. IDO and transplantation .................................................................................. 46 1.3.7. IDO and autoimmunity ................................................................................... 47 ! "###!1.4. HYPOTHESIS AND SPECIFIC AIMS ............................................................. 48 !!Chapter 2: MATERIALS AND METHODS ..................................................... 51 ! 2.1 ETHICS STATEMENT ....................................................................................... 52 2.2 ANIMALS .............................................................................................................. 52 2.3 MOUSE DERMAL FIBROBLAST ISOLATION, CULTURE AND TREATMENTS .......................................................................................................... 53 2.4 MOUSE PANCREATIC ISLET ISOLATION .................................................. 54 2.5 MOUSE SPLENOCYTE AND SPLENIC DENDRITIC CELL ISOLATION....................................................................................................................................... 55 2.6 GENERATION OF VIRAL VECTORS ............................................................ 55 2.7 TRANSDUCTION OF IDO GENE IN FIBROBLASTS .................................. 57 2.8 KYNURENINE ASSAY ....................................................................................... 58 2.9 WESTERN BLOT ANALYSES .......................................................................... 58 2.10 REVERSE-TRANSCRIPTASE ANALYSES .................................................. 59 2.11 PREPARATION OF 3D COLLAGEN AND CROSS-LINKED COLLAGEN MATRICES ................................................................................................................. 60 2.12 DETERMINATION OF ISLET VIABILITY .................................................. 62 2.13 ASSESSMENT OF ISLET FUNCTION .......................................................... 62 2.14 TRANSPLANTATION OF ISLET-FIBROBLAST COMPOSITE SCAFFOLDS ............................................................................................................... 63 2.15 HISTOLOGY ANALYSES ................................................................................ 64 2.16 EVALUATION OF FIBROBLASTS PROLIFERATION ............................. 65 2.17 VIABILITY AND IDO EXPRESSION OF TRANSDUCED FIBROBLASTS IN CCM ........................................................................................................................ 66 2.18 LASTING EFFECT OF IDO EXPRESSION IN COMPOSITE SCAFFOLDS ............................................................................................................... 66 2.19 ANTIGEN SPECIFIC PROLIFERATION ASSAY ....................................... 66 2.20 STATISTICAL ANALYSIS .............................................................................. 67 !! "$!Chapter 3: MECHANISM UNDERLYING DEFECTIVE IFN! -INDUCED IDO EXPRESSION IN NON-OBESE DIABETIC MOUSE FIBROBLASTS .......................................................................................................... 68 ! 3.1 INTRODUCTION ................................................................................................ 69 3.2 RESULTS .............................................................................................................. 70 3.2.1. IFN! fails to induce tryptophan catabolism and IDO expression in NOD splenic DCs and dermal fibroblasts. ......................................................................... 70 3.2.2. IFN! induces MHC I expression in NOD dermal fibroblasts. ........................ 75 3.2.3. IFN! reduces collagen expression in NOD dermal fibroblasts. ...................... 77 3.2.4. IDO gene-transduced NOD fibroblasts express IDO ...................................... 79 3.2.5. Defective STAT1 phosphorylation is responsible for impaired IFN!-induced tryptophan catabolism in NOD fibroblasts ............................................................... 83 3.2.6. LPS induces IDO expression in NOD dermal fibroblasts .............................. 85 3.3 DISCUSSION ........................................................................................................ 87 !Chapter 4: EMBEDDING ISLETS IN A LIQUID SCAFFOLD INCREASES ISLET VIABILITY AND FUNCTION ................................. 94 ! 4.1 INTRODUCTION ................................................................................................ 95 4.2 RESULT ................................................................................................................ 97 4.2.1. Cross-linked collagen matrix preserves normal morphology and insulin/glucagon expression in grafted islets. ........................................................... 97 4.2.2. Cross-linked collagen matrix improves the islet survival and viability. ......... 99 4.2.3. Cross-linked collagen matrix preserves islet insulin secretory function and reduces fibroblast proliferation rate. ....................................................................... 103 4.2.4. Length of IDO expression and effect on tryptophan catabolism in Lenti-IDO-transduced fibroblasts. ............................................................................................ 107 4.3 DISCUSSION ...................................................................................................... 109 !! $!Chapter 5: IMMUNOPROTECTION AND FUNCTIONAL IMPROVEMENT OF ALLOGENEIC ISLETS IN DIABETIC MICE, USING A STABLE IDO-PRODUCING SCAFFOLD ............................... 113 ! 5.1 INTRODUCTION .............................................................................................. 114 5.2 RESULT .............................................................................................................. 116 5.2.1. IDO-transduced fibroblasts embedded within CCM express IDO ............... 116 5.2.2 Local IDO expression suppresses proliferation of diabetic splenocytes co-cultured with allogeneic islets. ................................................................................ 118 5.2.3. Cross-linked collagen matrix improves viability and function of cultured murine islets. ........................................................................................................... 120 5.2.4. Local IDO expression delivered by lentiviral vector improves islet transplantation outcome. ......................................................................................... 123 5.2.5. Local expression of IDO improves insulin expression and reduces immune cell infiltration. ........................................................................................................ 125 5.2.6. Long-term expression of IDO in IDO-gene transduced islet grafts. ............. 129 5.3 DISCUSSION ...................................................................................................... 131 !!Chapter 6: DISCUSSION AND CONCLUSIONS ......................................... 135 ! 6.1 SUMMARY AND  DISCUSSION ..................................................................... 136 6.2 SIGNIFICANCE ................................................................................................. 143 6.3 FUTURE STUDIES ............................................................................................ 144 !!BIBLIOGRAPHY ......................................................................................................... 146 !  !!! "$!LIST OF TABLES  Table 2.1 Primers used for RT-PCR analyses. ................................................................. 60 Table 2.2 Primary antibodies used for immunostaining .................................................. 65 Table 2.3 Secondary antibodies used for immunostaining .............................................. 65 Table 4.1 Glucose-stimulated insulin secretion assay .................................................... 105                         ! "$$!LIST OF FIGURES  Figure 1.1 Transplanting pancreatic islets into the portal vein in humans ...................... 16 Figure 1.2 Factors influencing islet transplantation outcome .......................................... 26 Figure1.3 Kynurenine pathway of tryptophan catabolism in mammalian cells .............. 38 Figure 1.4 Proposed mechanism of IDO-mediated tolerance involving DCs and regulatory T cells .............................................................................................................. 44 Figure 3.1 Effect of IFN! on IDO expression in NOD splenic DCs and dermal fibroblasts. ......................................................................................................................... 72 Figure 3.2 Different effect of IFN! on IDO expression in dermal fibroblasts of C57BL/6 and prediabetic NOD mice. ............................................................................................... 74 Figure 3.3 MHC I mRNA expression in fibroblasts isolated from control and NOD mice............................................................................................................................................ 76 Figure 3.4 Type I collagen expression in dermal fibroblasts from C57BL/6 and NOD mice. .................................................................................................................................. 78 Figure 3.5 Tryptophan catabolism and microscopic evaluation of IDO expression in dermal fibroblasts following transduction with Ade-IDO vector. .................................... 80 Figure 3.6 IDO protein and mRNA expression in Ade-IDO transduced cells. ............... 82 Figure 3.7 IFN!-induced-STAT1 phosphorylation in C57BL/6 and NOD dermal fibroblasts. ......................................................................................................................... 84 Figure 3.8 LPS-induced IDO expression in C57BL/6 and NOD dermal fibroblasts. ...... 86 Figure 3.9 IFN!-mediated MHC I expression and type I collagen repression ................ 90 Figure 3.10 LPS induces IDO expression via the activity of JNK and independent of JAK/STAT1 signaling pathway ........................................................................................ 92 Figure 4.1 Histology assessment of the engrafted islets embedded within fibroblast populated collagen matrix (FPCM) or fibroblast populated cross-linked collagen matrix (FP-CCM) ......................................................................................................................... 98 Figure 4.2 Comparison of islet cell viability and survival within collagen and cross-linked collagen scaffolds .............................................................................................................. 101 ! "$$$!Figure 4.3 Islet cell viability and survival within cross-linked collagen scaffolds at different conditions ......................................................................................................................... 102 Figure 4.4 Islet insulin secretory function and fibroblast proliferation within collagen composite scaffolds.. ....................................................................................................... 106 Figure 4.5 Lasting effect of tryptophan catabolism in Lenti-IDO-transduced fibroblasts embedded within collagen scaffolds. .............................................................................. 108 Figure 5.1  IDO expression in transduced fibroblasts embedded within cross-linked collagen matrix (CCM) ................................................................................................... 117 Figure 5.2 IDO-expressing fibroblasts suppress the proliferation of alloantigen-stimulated mouse splenocytes. ........................................................................................ 119 Figure 5.3  Lenti-IDO fibroblast populated cross-linked collagen scaffold preserves islet insulin and glucagon expression while reducing caspase-3 expression in " cells in vitro......................................................................................................................................... 121 Figure 5.4 Islet grafts survival after transplantation ...................................................... 124 Figure 5.5 Evaluation of islet structure and immune cell infiltration in composite islet grafts ............................................................................................................................... 126 Figure 5.6 Characterization of intra-graft and –draining lymph nodes infiltrating immune cells ................................................................................................................................. 128 Figure 5.7 Evaluation of IDO expression in retrieved composite islet grafts ................ 130 Figure 6.1 Proposed model of IDO-mediated infectious tolerance ............................... 142             ! $"#!LIST OF ABBREVIATIONS    1-MT 1-Methyl tryptophan  2D Two dimensional culture Ade-IDO Adenoviral vectors carrying green florescent protein (GFP) plus human IDO gene Ade-Vect Adenoviral vectors carrying GFP alone  AIRE Autoimmune regulator  APC Antigen presenting cells ATP Adenosine triphosphate BMI Body mass index CCL22 C-C motif chemokine 22 C/EBP" CCAAT/enhancer-binding protein "  CM Collagen matrix CCM Cross-linked collagen matrix CFSE 5-(and 6)-carboxyfluorescein diacetate succinimidyl ester  COL-I Type I collagen CITR Collaborative Islet Transplant Registry CTL Cytotoxic T lymphocyte CTLA-4 Cytotoxic T lymphocyte associated protein-4 DCs Dendritic cells DCCT Diabetes Control and Complications Trial DMEM Dulbecco’s Modified Eagle Medium DPP-IV Dipeptidyl peptidase-IV ECM Extracellular matrix ER Endoplasmic reticulum ERK Extracellular signal-regulated kinase  FBS Fetal bovine serum FGF Fibroblast growth factor  FOXP3 Forkhead box P3 ! $#!FP-CM Fibroblast populated collagen matrix FP-CCM Fibroblast populated cross-linked collagen matrix  GAD65 Glutamic acid decarboxylase 65 GAPDH Glyceraldehyde-3-phosphate dehydrogenase  GAS !-activating sequences GCN2 General control nonderepressible 2 GFP Green florescent protein GLP-1 Glucagone-like peptide-1 H&E Hematoxylin and Eosin HGF Hepatocyte growth factor  HLA Human leukocyte antigen IA-2 Islet associated antigen-2 IAPP Islet amyloid polypeptide  IBMIR Instant blood mediated inflammatory reaction IDO Indoleamine 2,3- dioxygenase IFN Interferon IL Interleukin ISREs Interferon stimulatory response elements  JAK/STAT1 Janus kinase/signal transducer and activator of transcription 1  JNK c-Jun-N-terminal kinase Lenti-IDO Lentiviral vector carrying human IDO gene and Blasticidin resistance gene Lenti-Vect Lentiviral vector carrying Blasticidin resistance gene alone LFA Lymphocyte function associated antigen LPS Lipopolysaccharides KRBB Krebs-Ringer bicarbonate buffer  NK Natural killer MHC I Major histocompatibility class I MHC II Majjor histocompatibility class II MOI Multiplicity of infection NAD+ Nicotinamide adenine dinucleotide (NAD+) ! $"#!NF-#B Nuclear factor-#B NKG2D NK cell activation receptor NKT Natural killer T cells NOD Non-obese diabetic PBMC Peripheral blood mononuclear cell  PVA Polyvinyl alcohol  RT-PCR Reverse transcriptase-polymerase chain reaction SD Standard deviation  SEM Standard error of the mean STZ Streptozotocin TCR T cell receptor TDO Tryptophan 2,3- dioxygenase  TGF" Transforming growth factor " Th T helper TLR Toll-like receptor TNF Tumor necrosis factor Tregs Regulatory T cells tRNA Transfer RNA VEGF Vascular endothelial growth factor ZnT8 Zinc transporter 8           ! $"##!ACKNOWLEDGEMENTS This dissertation would not have been possible without the help of many people in different ways. I would like to express my deepest gratitude specially to the following:  I would like to express my sincere appreciation to my supervisor, Dr. Aziz Ghahary who always believed in me, was abundantly helpful and offered invaluable support and guidance. His wisdom, knowledge and commitment to the highest standards inspired and motivated me. Deepest gratitude is also due to the members of my supervisory committee, Dr. Bruce Verchere, Dr. Vincent Duronio and Dr. Lucy Marzban who were always there for me, despite their many academic and professional commitments. I truly appreciate their endless support, assistance and words of encouragement.  Many friends have helped me stay sane through this journey. Their wholehearted support and care helped me overcome setbacks and stay focused on my goals. I greatly value their honest friendship and deeply appreciate their belief in me. My deepest appreciation goes to Mr. Reza Khorasani and Dr. Kevin Aminzadeh. I wish to extend my appreciation to all my brilliant friends and colleagues at Ghahary Lab, from whom I learnt a lot, for their significant support and help during these years. I am also very grateful to my friends at iCORD for their constant kindness and support.  I wish to deeply thank the following funders for their support during my course of studies: Vanier Canada Graduate Scholarships program, the University of British Columbia, CIHR/Skin Research Training program, Transplantation Training Program, iCORD, and the Centre for Human Islet Transplantation and Beta-cell Regeneration.  I would like to acknowledge the academic and technical support of the University of British Columbia, and the Department of Medicine, specially the Experimental Medicine Program. Special thanks goes to Ms. Cornelia Reichelsdorfer, the ultimate problem solver, who was always there to help no matter when it was.  Most importantly, none of this would have been possible without the endless love, support and patience of my beloved parents Mr. Hassan Hosseini-Tabatabaei and Mrs. Mahroo Asady, and siblings: Arezou, Ali, Bahareh and Amir. Words fail me to express my deepest feelings of gratitude and indebtedness.  I am grateful to my family who always stood by me.  ! $"###!DEDICATION       To the true meaning of love,   My beloved parents,     My dearest brothers and sisters,        My Family,  And my little angel, Jina.    ! "!!!!!!!!!!!! Chapter 1:  INTRODUCTION !!!!!!!!!! !! #!1.1 TYPE 1 DIABETES  1.1.1 Diabetes mellitus Diabetes mellitus refers to a group of chronic metabolic disorders, defined by hyperglycemia and characterized by impaired metabolism of carbohydrates, lipids and proteins. The history of diabetes reveals the important contributions of Greeks, Romans and Egyptians. Aretaeus, the famous Greek physician coined the Greek word “diabetes” to describe this disease, meaning siphon, describing it as the melting down flesh and limbs into urine [1].  Diabetes has been shown to be the main cause of end-stage renal failure [2], non-traumatic lower limb amputation [3] and visual loss [4]. It is now estimated that about 380 million people worldwide are affected by diabetes. Likely more than half a billion people will have diabetes or will be at risk for developing the disease by 2035 [5]. There are two main types of diabetes: in type 1 diabetes, the mass and function of insulin-producing pancreatic beta cells (! cell) is markedly reduced by autoimmunity, leading to absolute insulin deficiency. This type of diabetes generally develops in childhood or early adulthood [6] Among 2.2 million Canadians with diabetes, approximately 10% have type 1 diabetes [7]. Type 2 diabetes, the most common form of diabetes, is characterized by inadequate !-cell mass and function that can no longer compensate for insulin resistance. The reduced !-cell mass in type 2 diabetes likely results from increased !-cell death as well as decreased !-cell regeneration through replication and neogenesis [8]. It is believed that western life style –characterized by minimal physical activity and increased caloric intake - contributes to obesity and insulin resistance, and when superimposed upon genetic susceptibility, results in type 2 ! $!diabetes in millions of people [6, 8]. The incidence of type 1 diabetes varies among countries, with the highest prevalence in northern regions and a consistent rise (3–4% per year) in all countries. Both genetic and environmental factors trigger the development of diabetes [9]. The critical role of environmental triggers is primarily supported by findings that the concordance rate of type 1 diabetes in monozygotic twins is about 50%, indicating that genetics can only account for about half of T1D susceptibility [10, 11]. Although no single environmental factor has been clearly identified, a number of environmental triggers have been implicated, including viruses [12] and diet [13, 14]. The classic symptoms of untreated diabetes include polyuria, polydipsia, and unexplained weight loss. The criteria for the diagnosis of diabetes mellitus includes symptoms of diabetes plus random blood glucose concentration of "11 mM (200 mg/dl), or fasting blood glucose of " 7 mM (126 mg/dl) or a 2h post-load blood glucose of " 11mM (200 mg/dl) during an oral glucose tolerance test, or a glycosylated hemoglobin (HbA1c) value higher than 6.5% [6, 15, 16].  1.1.2 Pathogenesis of type 1 diabetes A healthy adult pancreas is composed of over a million islets, making up <3% of the total pancreas and containing 1 mg of insulin [17]. Evidence obtained from histological analysis of the pancreas from patients with type 1 diabetes confirms the presence of immunological activity limited to insulin producing cells, including infiltration by autoreactive T cells (insulitis) [18-20]. The infiltrating lymphocytes may be found in the islet periphery (peri-insulitis), or be diffuse and present throughout the islet parenchyma (intra-insulitis). Although insulitis is robust in the NOD mouse model, type 1 diabetic patients present with lesions that have a relatively mild number of infiltrating cells [21] and fraction of infiltrated islets [22]. Thus, although the NOD mouse is a ! %!widely used and well-established model of autoimmune diabetes, it likely does not perfectly model human type 1 diabetes. Other evidence for an autoimmune etiology for type 1 diabetes is derived from observations of co-occurrence of type 1 diabetes with other autoimmune diseases (including celiac disease) [23, 24], the presence of autoantibodies to islet cell proteins including insulin and glutamic decarboxylase 65 (GAD65), genetic association of diabetes with specific human leukocyte antigen (HLA) genes linked with disease risk or protection, and the presence of T cells specific for ! cell antigens including insulin, GAD [11, 18] and islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP) [25]. Despite our understanding of the autoimmune etiology of type 1 diabetes, the mechanisms underlying the cause of this disease are still not fully understood.  Many individuals who develop type 1 diabetes demonstrate islet autoantibodies even years before disease onset [26]. This observation has allowed definition of preclinical phases of disease. Detection of multiple islet autoantibodies in subjects with genetic susceptibility is a biomarker of type 1 diabetes autoimmunity and a reasonable predictor of disease. Indeed, individuals with more than two antibodies against ß cell autoantigens including insulin, zinc transporter 8 (ZnT8), islet-associated antigen 2 (IA-2) or GAD65, have a 80-90% to develop type 1 diabetes [19]. While the presence of autoantibodies can predict type 1 diabetes, it does not mean that disease is imminent and it can take more than 5 years for clinical manifestations to occur [19]. Islet cell autoantibodies may be a marker of underlying T cell mediated ß-cell damage [27]. Autoantibodies are considered as markers for autoimmunity rather than effector mechanisms of destruction. Unfortunately, although autoantibodies are reasonable predictors of the likelihood of eventually developing type 1 diabetes, prediction of the exact onset of type 1 diabetes is still not possible using these or other biomarkers in clinical practice [11]. ! &!1.1.2.1 Destruction of pancreatic ! cells Cells from both the innate and adaptive immune systems contribute to insulitis. Both human and murine studies suggest that the destruction of ! cells in autoimmune diabetes is a cell-mediated process that involves both CD4+ and CD8+ T cells along with macrophage [28-31]. It is thought that an unknown trigger leads these cells to invade the islet, resulting in a cascade of events ultimately leading to ß-cell destruction [32-34].  Macrophages, as an important component of the innate immune system, play an important role in antigen presentation and activation of !-cell specific-cytotoxic CD8+ T cells and generation of CD4+ autoreactive effector T cells. In addition, during early phases of type 1 diabetes, macrophages infiltrate the islets and when activated, generate reactive oxygen and nitrogen species along with pro-inflammatory cytokines and chemokines likely contributing directly to !-cell death and dysfunction. By secreting such molecules, macrophages can either initiate damage to nearby islet cells or recruit other immune cells to the islet area. Destruction of pancreatic ! cells is considered to be primarily T cell mediated.  ß-cell specific cytotoxic CD8+ T cells can directly kill ! cells. Effector CD4+ T cells, also known as T helper (Th) cells can differentiate into 4 subtypes: Th1, Th2, regulatory T cells (Tregs) and Th17 that cooperate and influence the outcome of transplanted graft [35]. CD4+ T cells can initiate the activation of B cells leading to antibody production [32]. In addition, both CD4+ and CD8+ T cells contribute to ! cell destruction by release of cytotoxic molecules including cytokines (e.g. interleukin (IL)-1, interferon (IFN) # and tumor necrosis factor (TNF) $, as well as the cytotoxic mediators granzyme B and perforin [36]. Pro-inflammatory and pro-apoptotic cytokines upregulate Fas on ß-cells, and increase the production of Fas ligand, free radicals and nitric oxide by T cells. ß-cell specific T cells induce !-cell apoptosis by activation of caspase pathways following Fas/Fas ! '!ligand interaction, nitric oxide and reactive oxygen species and disrupt ! cell membrane via perforin/granzyme B. Other pro-inflammatory cytokines include IL-2, IL-12, IL-17 and IL-18, which further enhance the inflammatory response [37].  The autoimmune process in type 1 diabetes is thought to be initiated by an as yet unknown environmental trigger factor. This trigger, possibly an enterovirus or other pathogen causing primary damage to ! cells [38], leads to activation of antigen-presenting cells (APCs) and exposure to !-cell autoantigens, either as soluble proteins or through engulfment of apoptotic ! cells in pancreatic lymph nodes. These !-cell autoantigens include insulin, GAD65, and IA-2, and ZnT8 [37]. When presented in the local lymph nodes by disease-risk HLA class I or II molecules on APCs, epitopes derived from these autoantigens prime autoreactive effector CD4+ T cells. Following activation, CD4+ T cells secrete cytokines that result in the induction of ! cell-specific cytotoxic CD8+ T cells. Activated CD4+ and CD8+ T cells migrate to the islets and mediate ! cell damage via secretion of pro-inflammatory cytokines, activation of apoptotic pathways or direct cytotoxic CD8+ -mediated lysis. ! cell death likely contributes to a vicious cycle of further inflammation, generation of more autoreactive T cells, and further ß-cell death [37, 39].  1.1.2.2 Dysregulation of immunotolerance Cells regulating peripheral immunotolerance are also thought to play an important role in the development of autoimmunity against ! cells. Tregs and natural killer T cells (NKT cells) are two important suppressor immune cell subtypes whose dysfunction may contribute to the development of type 1 diabetes [40]. It has been suggested that subjects with type 1 diabetes do not have the same number of Tregs as healthy individuals and moreover that these cells have ! (!decreased suppressive capabilities [40, 41]. The adoptive transfer of Tregs has shown promise in amelioration of the disease in the non-obese diabetic (NOD) mouse model of autoimmune diabetes [42] and the potential of this approach in humans is under investigation. Tregs are well-known regulators of immune system. Suppressing proliferation of effector cells and shutting off active T cell responses, Tregs are believed to play an important role in the prevention of autoimmunity. Furthermore, when NKT cells get activated they produce IL-4, which shifts the Th1 type response towards a Th2 response. The Th1 response is characterized by effector Th1 cells producing pro-inflammatory cytokines such as IFN# and lymphotoxin-$. In contrast, effector Th2 cells produce a different profile of cytokines including IL-4, IL-5, IL-9, IL-10, and IL-13). Th1 cells induce delayed-type hypersensitivity reactions and activate the antimicrobial defense for intracellular pathogens such as Mycobacteria. Th1 cytokine production is also characteristic of many organ-specific autoimmune diseases, including rheumatoid arthritis, type 1 diabetes, and others. On the other hand, Th2-produced cytokines stimulate B cell proliferation and differentiation into antibody-secreting plasma cells. Therefore, the Th2 response is more important for providing protection against certain extracellular pathogens and their secreted products, such as bacteria and a variety of parasites, and is also involved in asthmatic reactions [43-45]. Subjects with type 1 diabetes have been shown to have decreased populations of NKT cells and reduced ability to produce and secrete IL-4 [46]. In those who develop aggressive autoimmunity, the possible cell mediatory regulatory pathways including IL-10+ islet-specific Tregs that can kill APCs, are insufficient to fully control disease [47]. 1.1.2.3 Environmental factors The fact that in case of type 1 diabetes, disease susceptibility but not the disease itself is ! )!inherited, suggests that factors other than genetic susceptibility are involved in the development of type 1 diabetes. Although there is evidence in animal models that infections can trigger autoreactivity towards ! cells (e.g. molecular mimicry between viral and ! cell proteins, bystander activation of T cells, or impairment of Th1/Th2 balance following infection), there is no strong indication to support this hypothesis in humans. There are several reports of co-existence of viral infections and type 1 diabetes or virus specific RNA, DNA or antibodies in individuals with recent onset of type 1 diabetes [32]. It has been proposed for several years that enteroviruses are closely linked with type 1 diabetes. There are case reports indicating fulminant or sudden onset of diabetes alongwith high antibody titers for coxsackie virus B4 [48]. Other suggested viruses include rubella virus and cytomegalovirus [49]. Pathological situations that expose autoantigens and result in their abnormal presentation by APCs remain elusive. Dietary factors like gluten consumption and cow’s milk have been also investigated. Although animal studies suggest that gluten-free diet prevents the development of type 1 diabetes [50, 51], there is no evidence in humans yet. Furthermore, there are data on cow’s milk consumption and development of diabetes in animal models, and also observations on protective effects of prolonged breast-feeding against type 1 diabetes in susceptible infants. A preliminary report of the international randomized clinical trial of TRIGR indicates that supplementing breast milk with highly hydrolyzed milk formula may decrease the development of islet autoantibodies [14]. Significant vitamin D deficiency [52] and genetic variation in vitamin D gene [53] are linked with predisposition to type 1 diabetes Furthermore, it has been suggested that caesarean section and low birth weight increase the risk of type 1 diabetes. Early life events seem to be of importance in determination of genetic-environment contribution leading to type 1 diabetes [13].    ! *!1.1.2.4 Genetic factors Considering the multiplicity of genes involved, it is possible that type 1 diabetes is a heterogeneous multifactorial disease. Over 50 loci have been identified to be associated with type 1 diabetes. The most important genetic influence on susceptibility to type 1 diabetes comes from HLA II genes, accounting for about 40-50% of the genetic contribution [10, 11]. Alleles presenting an aspartic acid at the residue 57 of the DQ! (DQ6/DQB1*0602) represent dominant protection while those carrying serine, alanine or valine residues at this site (DQ!8/DQB1*0302 on DR4 haplotypes or DQ2/DQB1*0201 on DR3 haplotypes) is associated with susceptibility to the disease [10]. The other non-HLA loci include the preproinsulin gene, cytotoxic T lymphocyte associated protein-4 (CTLA-4), protein tyrosine phosphatase-22 and IL-2 receptor $ [10]. Mutations in critical immunoregulatory genes like the autoimmune regulator (AIRE) [54] or forkhead box P3 (FOXP3) [55] can result in complex immune disorders including type 1 diabetes.   1.1.3 Current treatments A critical and transformative discovery in the history of diabetes came in 1920s. Before the discovery of insulin, fourteen-year-old Leonard Thompson was near death when he was admitted to hospital in January 1922. A child with type 1 diabetes, his weight had dropped to 60 pounds when Dr. Frederick Banting injected him with a pancreatic extract isolated following the research efforts of Drs. Banting, Charles Best, James Collip, and J.J.R.McLeod. Later, Thompson received refined formulas that allowed him to live with the disease for another 13 years until he died of pneumonia ! "+!Insulin therapy, still the conventional treatment for type 1 diabetes almost 100 years following its discovery, has significantly decreased the mortality and morbidity of type 1 diabetes. As confirmed by the Diabetes Control and Complications Trial (DCCT), tight glucose control prevents and even reverses several long-term complications of type 1 diabetes [56]. Intensive insulin therapy requires multiple daily injections or continuous subcutaneous insulin infusion pump and is a relatively an effective method of maintaining glycemia and Hb A1c levels in the normal range [57]. Although insulin therapy prevents can ameliorate hyperglycemia and acute sequelae like ketoacidosis and coma, it fails to completely prevent the secondary complications of type 1 diabetes because even with meticulous insulin therapy, blood glucose control is imperfect.. In addition, intensive insulin therapy carries with it significant risk of hypoglycemic episodes, which can be lethal [57]. These complications due to lifelong hyperglycemia include micro- and macro-vascular disease that can lead to kidney failure, blindness, and amputations [58]. Moreover, intensive insulin therapy comes with the risk of hypoglycemic events [59, 60] that can lead to seizure [35], coma [61] and even death. These life-threatening events and the devastating and complications associated with diabetes severely affects the quality of life of individuals with type 1 diabetes, and additionally imposes a considerable burden on their family and society. Improved strategies for management of type 1 diabetes are clearly still needed [62].  Whole organ pancreas and pancreatic islet transplantation are the currently available forms of ! cell replacement in humans, though islet transplantation remains an experimental procedure available to only a few of the millions of persons with type 1 diabetes. Both strategies have the advantage of resulting in restoration of endogenous insulin production in response to blood glucose concentrations [7, 35, 49]. Replacing ! cells via transplantation can provide good ! ""!glycemic control and may prevent or stabilizing diabetic complications [63], but at the price of permanent immunosuppression to prevent rejection of the transplanted cells. Islet transplantation may be a more attractive procedure compared to the whole organ transplantation in select individuals, as the procedure offers the advantage of using a small volume of infused cells, making the procedure minimally invasive as well as potentially decreasing the amount of immunosuppression required. Furthermore, islet transplantation is generally considered as a day procedure, without requirement for general anesthesia, major surgery, or prolonged hospitalization as is the case with whole pancreas [7, 64]. Nonetheless, results with whole pancreas organ transplantation in type 1 diabetes, usually transplanted with kidney, are as good or better as islet transplantation in terms of organ survival and glycemic control.   1.1.4 Animal models of insulin-dependent diabetes mellitus Animal models of diabetes, especially those in which diabetes develops spontaneously from autoimmunity, have contributed extensively to our current understanding of type 1 diabetes pathogenesis and the development of treatment strategies. Several animal models of diabetes have been generated by chemical, surgical (pancreatectomy) and genetic manipulations in different species. Streptozotocin (STZ) is a nitrosourea derivative isolated from fermentations of Streptomyces achromogenes that is classically an anti-neoplastic agent and antibiotic [45, 46]. It has been shown that a single dose of STZ can induce rapid beta cell destruction and permanent hyperglycemia in rodents. Due to its similar structure to glucose, STZ is internalized into the beta cell via the GLUT2 glucose transporter [65]. The cytotoxicity of STZ is due to its strong alkylating properties causing multiple DNA breakages, necrosis and insulin deficiency [65, 66]. ! "#!The multiple low-dose STZ model has been used extensively to study the immunological pathways that lead to insulitis and !-cell death. In contrast to spontaneous rodent models of type 1 diabetes, however, STZ-induced diabetes does not require autoimmunity for induction of insulin-dependent diabetes. STZ-induced diabetes is therefore not considered to be an ideal animal model of type 1 diabetes [65], but is suitable for studies of islet allograft rejection in diabetic recipients  The NOD mouse is a well-studied spontaneous animal model of type 1 diabetes that was developed by selectively breeding offspring from JcI-ICR mice, a strain initially generated for the study of cataracts. Early histological evidence showed infiltration of islets by macrophages and dendritic cells (DCs) around 3 weeks of age [11]. NOD mice develop insulitis at 4–5 weeks of age, followed by subclinical T cell-mediated destruction of insulin producing cells.  Absolute diabetes typically occurs when mice are between 12 and 30 weeks old. Although the NOD strain shows several similarities to human type 1 diabetes, unlike in human subjects, ketoacidosis is relatively mild and diabetic NOD mice can survive for weeks without receiving insulin. Also, in contrast to humans, there is a significant gender difference with 80-90% of females, but less than 60% of males developing diabetes in most colonies. The onset of disease is also faster in female compared to male NOD mice.  The BB rat was generated in Bio-Breeding Laboratories in 1974. This strain develops diabetes with the same incidence in males and females [67] around 12 weeks of age [65]. As in human type 1 diabetes, ketoacidosis is serious and requires regular administration of insulin for survival. As in NOD mice, T cells, B cells, macrophages and natural killer (NK) cells infiltrate the islets and contribute to development of diabetes. Unlike humans, BB rats are lymphopenic and lack a key regulatory T cell subset that express RT6 [67]. Furthermore, they are severely ! "# !deficient in CD8+ T cells due to a genetic defect. Nonetheless, the BB rat is a useful model of autoimmune diabetes, though much less widely used than the NOD mouse.   1.2 ISLET TRANSPLANTATION Transplantation of pancreatic insulin-producing cells, also known as islet transplantation, is a promising approach that could obviate the need for daily insulin injections. This procedure requires the isolation of pancreatic islets from organ donors and their infusion into the hepatic portal vein of the recipient. Islet transplantation trials began with some success in the early 1990s [68, 69], raising hope that this procedure would become the definitive treatment for restoring insulin production and efficient blood glucose regulation. Unfortunately, international clinical trials of islet transplantation in type 1 diabetes have demonstrated a constant decline of islet graft function post-transplantation and therefore a low rate of long-term successful engraftment [70]. Currently, human islet transplantation is limited by the shortage of available islet donors and poor survival of the islets after transplantation [7, 35]. 1.2.1 Islet isolation Pancreatic islets compose a small portion (<3%) of the whole pancreas; the isolation process aims to remove the exocrine tissue through enzymatic digestion of the extracellular matrix (ECM) and mechanical separation followed by density gradient purification [71]. After pancreas dissection from a cadaveric organ donor, the pancreas is normally maintained in a chilled preservation solution. University of Wisconsin solution alone or plus perflurocarbon (the two-layer method) are most often used for organ preservation [72] although it has been reported that ! "%!the two-layer method does not provide significant beneficial effects on islet isolation and transplant outcomes [73]. Since collagenase for exocrine pancreas digestion is delivered via the pancreatic duct, pancreatic ductal preservation seems critical [74]. Following arrival at the islet isolation facility, the pancreas is distended with a controlled collagenase perfusion and enzymatically digested in a Ricordi chamber [75]. This chamber is specifically designed for effective pancreas digestion and gentle mechanical dissociation with accurate temperature control that effectively collects aliquots of the digested pancreatic tissue at different time points with a large volume of solution. The final step is purification of isolated islets from the digested exocrine tissue, an important step for achievement of high islet yields. Islet purification is commonly performed using density gradient centrifugation [76, 77]. Subsequently, islets are washed and put into a transplantation bag or culture medium before transplantation.  1.2.2 Current status of islet transplantation Lacy and Kostianovsky in 1967 were the first to develop a novel collagenase-based method for successful isolation of the islets from rat pancreas [59]. Successful reversal of type 1 diabetes by experimental islet transplantation was later reported by Ballinger and Lacy in 1972 [78] and shortly after by Reckard and Barker in 1973 [79], initiating a new era with great potential for the treatment of type 1 diabetes. The current islet transplantation procedure involves the delivery of purified islets to the liver via the portal vein (Figure 1.1). Starting in March 1999, a group of scientists and clinicians at the University of Alberta performed a milestone trial of clinical islet transplantation demonstrating successful reversal of diabetes in seven consecutive type 1 diabetes patients [80]. The so-called “Edmonton Protocol” of islet transplantation was a significant breakthrough in this ! "&!field. This protocol includes multiple infusions of isolated islets from multiple donors, immediate transplantation without culture of the islets prior to transplant, and steroid-free immunosuppression. The one-year follow up of these seven patients demonstrated 100% insulin independence, less fluctuation of blood glucose levels and no episodes of severe hyperglycemia. Data obtained in a trial reported by the Edmonton group in 2004 of a 5-year follow up of 65 islet transplant recipients, showed that the insulin-free status was maintained in less than 10% of transplant recipients. However, more than 80% of these recipients still demonstrated at least partial islet graft function and had lower insulin requirements to maintain blood glucose control [81]. Despite the gradual loss of complete insulin independence, the Edmonton group has suggested that since an important goal is to prevent hypoglycemia unawareness and tragic “death-in-bed” events, even partial islet graft survival might suffice [82].   ! "'!   Figure 1.1 Transplanting pancreatic islets into the portal vein in humans. [Figure adapted from Ref. 83].  Following digestion of pancreatic tissue in the Ricordi chamber, pancreatic islets are purified and separated from exocrine tissue and debris.  Generally the transplant procedure involves percutaneous cannulation of a branch of the portal vein followed by gravity infusion of the islet preparation [84].   1.2.3 Transplantation immunology 1.2.3.1 Antigen recognition pathways Since the discovery of major histocompatibility (MHC) antigens in 1967 [85], the field of organ and tissue transplantation has developed remarkably. Positive outcomes of graft acceptance have Pancreatic Islet Transplantation ! "(!been achieved by matching donor and recipient for MHC antigens. Several components of the immune system play roles in tolerance or graft rejection such as antibodies, APCs, helper and cytotoxic T cells, surface molecules of immune cells, signaling mechanisms, and cytokines [86]. Compared to HLA compatibility, ABO compatibility is of much less importance in graft survival, although ABO incompatibility can lead to hyper-acute rejection of primarily vascularized grafts [86].   Like any other immune response to a foreign antigen, the allo-immune response occurs in two broad phases: the innate and adaptive phases [27]. The innate immune system acts as the “primitive line of defense” against foreign antigens, whether an allograft or an invading microbial pathogen. The cellular components of this phase of the immune system are APCs such as DCs and macrophages, NK cells and neutrophils that are non-antigen-specifically activated by local inflammation. Toll like receptors (TLR), the pathogen recognition receptors on APCs, initiate a signaling pathway resulting in translocation of nuclear factor- %B (NF-%B) and the expression of pro-inflammatory cytokines (e.g. TNF$, IL-1, 6, 8 and 12) [87]. In the innate phase allo-immune response, APCs are activated due to inflammation in the transplanted organ. Pro-inflammatory cytokines induce DC maturation and migration to secondary lymphoid organs and expression of surface proteins necessary for T cell stimulation such as CD40, CD80 and CD86 [87].   In secondary lymphoid organs, DCs activate naïve T cells and initiate the adaptive phase of the immune response. Following presentation of alloantigens to CD4+ (helper) and CD8+  (cytotoxic) T cells, the “expansion phase” begins which results in logarithmical expansion of activated T cells. Subsequently, T cells gain effector functions that allow them, with the help of B cells and other mononuclear cells, to eliminate the foreign antigen and reject the graft, known ! ")!as the “effector phase”. Allorejection is directed most commonly via cell-mediated reactions, such as delayed-type hypersensitivity and cytotoxic T lymphocyte (CTL)-mediated cytotoxicity, and less commonly through antibody-dependent cellular cytotoxicity and antibody/complement mediated lysis [87].  Two main recognition pathways have been suggested for generation of alloreactive T cells. In the “direct recognition” pathway, donor APCs migrate from the graft to local lymph nodes, presenting alloantigens to the recipient’s alloreactive T cells and activating them [27]. This mechanism is thought to be important in early acute rejection. In the “indirect recognition” pathway, peptides derived from the processing of alloantigens by host (recipient) APCs result in T cell activation. This pathway is thought to be involved in initiation and perpetuation of allograft rejection [27].  1.2.3.2 Hyperacute, acute and chronic rejections Antibodies play the main role in the hyperacute graft rejection of primarily vascularized organs, and are detectable in graft recipients in high titers following rejection. Hyperacute rejection is a complement-dependent mechanism characterized by vessel thrombosis leading to graft necrosis, and can occur within minutes post-transplantation in recipients with pre-existing anti-donor antibodies (e.g. ABO blood type antibodies) [27]. Hyperacute allorejection is avoided in most cases by checking for ABO compatibility, and by cross-matching techniques between donor graft cells and recipient sera to exclude the presence of anti-donor HLA antibodies in the recipient. Hyperacute rejection is also typically observed in xenotransplantation of vascularized organs when no immunosuppression is provided [87].  The process of acute rejection usually initiates between one week to several months after ! "*!transplantation [87] and is due to mismatched HLA molecules present on all cells of the body [27]. Whereas acute rejection is mainly a T cell-dependent process, allograft destruction results from a wide-ranging selection of effector immune mechanisms. Chronic organ rejection, happening months after surgery, is caused by inflammatory vascular injury to the graft such as atherosclerosis of the graft vessels or glomerular/tubular fibrosis in renal transplantation. The mechanism underlying chronic rejection includes alloreactivity or other factors including but not limited to ischemia-reperfusion, toxicity of immunosuppressive drugs, and infections. Histological analysis of rejected organs has demonstrated dense infiltration by macrophages and features of fibrosis and scarring [27].  1.2.3.3 Role of antibodies, leukocytes and cytokines The process of hyperacute rejection starts with alloantibodies combining with HLA antigens on endothelial cells, followed by complement fixation and accumulation of polymorphonuclear cells. Enzymes released from polymorphonuclear leukocytes result in endothelial damage. Subsequent accumulation of platelets causes development of thrombosis and graft dysfunction.  Activated alloreactive CD8+ T cells release granzyme B, perforin, and toxic cytokines such as TNF$, contributing to cytotoxic lysis of transplanted tissue. CD4+ helper T cells primarily produce various types of cytokines. Furthermore, CD4+ helper T cells can differentiate into one of several subtypes, including Th1, Th2, Th17, or Tregs, further defining the type of immune response [88]. Indeed, the identification of pro-inflammatory Th17 effector cells and Tregs has improved our understanding of graft rejection or tolerance.  Pro-inflammatory cytokines and chemokines released from macrophages and primed helper T cells recruit donor-specific CD4+T cells, CD8+ cytotoxic T cells, and B cells along ! #+!with nonspecific inflammatory cells, which constitute the majority of cells infiltrating an allograft [87]. Cytokines secreted from T cells also activate macrophages and other inflammatory leukocytes as well as induce upregulation of HLA molecules on graft cells [86]. Activated T cells also stimulate B cells to produce anti-graft antibodies. It is worth mentioning that the involvement of B cells in alloimmune rejection is not limited to the production of alloimmune antibodies but also antigen presentation and secretion of pro-inflammatory cytokines [86, 87]. All of these cellular and humoral factors likely contribute to the rejection process that destroys the graft. 1.2.4 Limitations of islet transplantation Early islet transplantation trials raised hope that this procedure would become a widely available treatment for restoring efficient blood glucose regulation. Unfortunately, the international trials conducted around the world have demonstrated a constant decline of islet graft function post-transplantation and therefore a low rate of successful engraftment [70]. Two of the most important limitations of human islet transplantation are the currently inadequate means for improving islet survival post-transplantation, and the limited supply of donor islets [7, 35]. The Collaborative Islet Transplant Registry (CITR) collects and analyses islet transplant data from all transplant centers in the United States and Canada, as well as from most centers in Europe and Australia, in order to identify critical factors associated with islet transplant success and thereby help to improve this procedure. The most recent CITR report included data on 571 allogeneic islet transplant recipients, who received a total of 1,072 infusions from 1,187 islet donors [88]. The data show that only 31% of islet transplant recipients received a single infusion of islets, while 47% received 2 infusions, 20% received 3 infusions and 2% received 2-6 infusions [88]. ! #"!Considering that the 5-year insulin independence rate after islet transplantation is less than 10%, it is critical to understand and improve the factors affecting islet function and survival and consequently the transplantation success rate. It is believed that both immune and non-immune factors are involved in reduction of ! cell mass and graft failure [71]. 1.2.5 Factors affecting transplantation outcome 1.2.5.1 Donor variables Due to poor islet isolation efficiency, more than two donors are required to achieve the critical islet mass for successful islet transplantation. One of the factors having a substantial effect on the outcome of islet transplantation is the quality of the donor pancreas. Islet yield and function are affected by several donor factors including but not limited to body mass index (BMI) [89], age [90], stability of the brain-dead donor(s)[76] and history of pancreatitis or pancreas damage [91].  Pancreata obtained from obese donors result in a better islet yield versus those isolated from lean donors [89]. Studies performed at University of Minnesota have suggested that successful islet transplantation would be difficult to achieve with low-BMI donors. Although older pancreas donors were previously considered to be better candidates for islet isolation because of the higher islet yield, today it is believed that islets from younger donors show better insulin secretory function and that modifying isolation methods could result in better islet yield from such donors [90]. Unstable circulation/blood pressure and cardiac arrest are among the exclusion criteria for selection of suitable donor candidates, therefore stable status of brain-dead organ donors is considered as an important factor [76]. Isolation of islets from donors with a history of type 2 diabetes is less desirable since such islets already have functional defects [92]. ! ##!Fibrotic pancreata caused by chronic pancreatitis usually result in very poor islet isolation, as such organs are resistant to digestion and the purity of isolated islets is remarkably lower [91]. Finally, organ procurement plays an important role in successful islet isolation. Pancreata that are procured locally give better islet yields than those that are procured distantly and transported, because of shorter duration of cold storage [71, 93].  1.2.5.2 Disruption of the islet extracellular matrix The ECM of adult human islets includes an incomplete capsule surrounding the islets that consists of a monolayer of fibroblasts and the collagen fibers that they produce. In close proximity to the capsule there exists a peri-insular basement membrane [94]. Loss of islet peripheral ECM and perinsular basal membrane during the isolation process affects islet engraftment potential. Isolation interrupts the islet vasculature and innervation and also disconnects !-cells from their interaction with macromolecules of the ECM. It is believed that such interactions are crucial for ! cell survival and function by preserving specific intercellular relationships [94]. In parallel, attachment of peripheral islet cells to ECM benefits ! cells by maintaining islet architecture. Although the true implications of islet-ECM interactions are not well understood, it is likely that they significantly affect transplantation outcomes [94]. The role of ECM is discussed in more detail later in this thesis.  1.2.5.3 Hypoxia Islets receive 15-20% of the blood circulation of the pancreas in their native environment even though they compose only 1-2% of the whole pancreas [8]. Transplantation of islets into the liver ! #$!requires infusion of islets into the portal vein. When isolated, islets go through multi-hour hypoxia (because they are not perfused as in situ) and then are trapped within capillaries in the liver where they receive oxygen and nutrients by diffusion until revascularization from the hepatic arterial system takes place [95]. The process of revascularization takes days to weeks and even in established islet grafts the oxygen pressure is less than native islets of pancreas. In large islets, more cell death and less insulin secretion is seen in the ! cells located at the core, due to poor oxygenation, and nutrient and glucose delivery to this site, and activation of apoptotic pathways [71].    1.2.5.4 Innervation The sympathetic and parasympathetic fibers of the autonomic nervous system highly innervate the islets and regulate the secretion of islet hormones. The islet autonomic nervous system is of physiological importance in mediating the cephalic phase of insulin secretion.  It is also involved in synchronising the islets allowing oscillations of islet hormone secretion, and in regulating islet hormone secretion during metabolic stress like hypoglycemia [96]. It is shown that sympathetic innervation is required for islet cell-cell interaction [97] and inhibits basal and glucose-stimulated insulin secretion [96]. Parasympathetic innervation of the islet is derived from the vagus nerve and is important for meal-stimulated insulin secretion [98]. Disruption of innervation due to the islet isolation process [94] is likely to significantly affect islet graft function, by interrupting autonomic regulation of insulin and glucagon secretion.   ! #%!1.2.5.5 Instant blood-mediated inflammatory reaction One of the main reasons for necessity of multiple islet infusions is that 50-60% of islets are lost within the first few days after transplantation. The process may start within minutes after portal infusion of the islets and is an innate immune response known as instant blood-mediated inflammatory reaction (IBMIR). The intraportal injection of islets is accompanied by a vascular injury that activates coagulation pathways and the complement system resulting in inflammation, ischemia, and activation of apoptotic pathways in ß cells [71]. Nonetheless, further work needs to be done to full understand the rapid loss of islets following transplant, and to develop ways to ameliorate IBMIR.  1.2.5.6 Glucotoxicity Diabetes is a metabolic disorder with multiple defects in glucose, lipid and protein metabolism. Following transplantation and prior to establishment of the islet graft, ! cells encounter an environment of hyperglycemia and increased secretory stress in the recipient. This environment can lead to impaired ß-cell function, inflammation and ß-cell apoptosis [99]. Increased insulin demand and secretion may lead to endoplasmic reticulum (ER) stress, resulting in loss of ß cell function and cell death. In addition, as discussed below, hyperglycemia may be a driver of amyloid formation in islet transplants, also contributing to ! cell dysfunction and death [100]. High circulating levels of free fatty acids as well as cholesterol accumulation likely contribute to impaired ß cell function in type 2 diabetes, and may therefore also contribute to graft dysfunction [101]. The deleterious effects of fatty acids on ! cells are seen in the presence of hyperglycemia (so-called glucolipotoxicity) [101].  ! #&!1.2.5.7 Endoplasmic reticulum stress Like all secreted proteins, the early steps of proinsulin biosynthesis and post-translational modification occur in the ER. ! cells have a highly developed and active ER. ER stress occurs due to an imbalance between client protein load and folding capacity of the ER [102]. The role of ER stress in association with ! cell death during islet isolation and transplantation has been previously investigated [103]. It has been shown that the unfolded protein response gets activated immediately post-transplantation while islets are recovering from inflammation and injury due to isolation and implantation and are more susceptible to harmful effects of ER stress [103]. In such conditions, even mild hyperglycemia can activate the unfolded protein response- mediated apoptotic pathways [102, 103] and may contribute to islet graft dysfunction and loss.  ! #'!   Figure 1.2 Factors influencing islet transplantation outcome. [Figure adapted from Ref. 104]. Transplanted islets encounter numerous overlapping forces all conspiring to limit islet graft function and/or survival. Other factors not shown in this figure include islet amyloid deposition and loss of islet ECM during isolation process. Addressing each of these factors could contribute to improving islet transplantation outcome and developing strategies for widespread clinical application of this procedure.  ! #(!1.2.5.8 Islet amyloid formation Islet amyloid polypeptide (IAPP) is a 37 amino acid polypeptide co-secreted with insulin from ß cell secretory granules. Human IAPP is able to aggregate to form fibrils. Accumulation of IAPP is associated with ! cell death [105]: amyloid fibrils interact with ! cells and have been shown to induce apoptosis. IAPP aggregates also attract and activate macrophages to secrete pro-inflammatory cytokines, leading to ! cell dysfunction [106]. Furthermore, it appears that small amylin fibrils and toxic oligomers can damage cell membrane by generating ion-permeable pores leading to disruption of ionic-homeostasis and potentially ! cell apoptosis [105]. There is evidence indicating rapid formation of amyloid deposits in transplanted human islets [107, 108]. In agreement with other studies [92, 109, 110], our research group has shown the presence of amyloid in cultured human islets [111].  It is still unclear why despite elevated amylin secretion in both obesity and diabetes, amyloid deposits are more common in islets of diabetic vs. obese individuals [8].  1.2.5.9 Immunosuppressive drugs Immunosuppression after allogeneic islet transplantation is required to prevent both graft rejection and recurrent autoimmunity. In addition to a number of potential side effects of the immunosuppressive drugs used to prevent islet graft rejection, some of these drugs are also toxic to ! cells [84]. For example, corticosteroids increase blood glucose levels and are known to directly impair ! cell function [112]. Indeed, omission of corticosteroids in the Edmonton protocol is thought to be a key reason why this protocol resulted in improved islet transplant outcomes. Sirolimus (rapamycin), an mTOR inhibitor, has been shown to impair islet ! #)!engraftment [113], angiogenesis [114] and inhibit ! cell replication [115], while inducing insulin resistance [116]. There is evidence that tacrolimus (a calcineurin inhibitor) and micophenolate mofetile (an inhibitor of inosine monophosphate dehydrogenase) decrease both insulin transcription and translocation [117], whereas micophenolate mofetile is shown to inhibit ! cell neogenesis as well [118].  1.2.5.10 Immune rejection Immune attack clearly plays a critical role in graft loss. Allogeneic islets trigger immune mediated rejection which immunosuppression drugs are used to counter. Multiple infusions of islets likely sensitize the recipient against islet donor tissue antigens and increases the risk of rejection following further infusions. Moreover, it is likely that recurrent autoimmunity occurs in islet transplant recipients, which can adversely affect islet allograft survival [71]. Unlike other organ transplants, the immune system of type 1 diabetic recipients is already primed to attack transplanted !-cells by recurrent autoimmunity. Indeed, in recipients who share HLA class 1 alleles with islet donors, recurrent anti-! cell-specific CD8+ T cell–mediated reactivity associated with loss of islet allograft function has been observed [119]. 1.2.5.11 Transplantation site An optimal transplantation site for islets should be highly vascularized, having a suitable blood flow and oxygen pressure, accessible and providing an ECM for the islets similar to their native environment in the pancreas [94]. Many researchers have suggested that transplantation of islets into the liver via portal vein is not the best option, but there is no agreement as to what the ! #*!optimal site is as yet [71]. Other potential sites for islet transplantation include, but are not limited to, the kidney capsule, spleen, peritoneal cavity, testis, omentum, muscle and pancreas [84]. These sites have been tested in animal models with some success. Transplanting islets into the pancreas would appear to be an attractive option, but due to the risk of pancreatitis and other complications, this site is less likely to be a feasible option in humans. 1.2.6 Current strategies to improve islet transplant outcome 1.2.6.1 Addressing the shortage of islet donors  By 2012, over 750 islet transplants had been performed in more than 30 international centres, demonstrating that this procedure is close to becoming a well-recognized clinical therapy rather than just an experimental treatment [95]. Initial reports of the Edmonton Protocol in 2000 clearly suggested that at least 2 and up to 4 donors were required in each recipient to achieve insulin independence [70]. Considering the ratio of supply and demand, it is truly important that we address the shortage of islet donors and establish strategies to improve isolation and early survival of islets.    Generally, pancreatic islets are provided from pancreata retrieved from heart-beating donors, brain dead donors, or organ donors that have succumbed after cardiac death (also known as DCD). Currently, less than 0.1% of persons with T1D can receive islet transplants each year, given the shortage of available donor pancreases [104]. Public education and encouraging organ donation is critical to address the issue of organ shortage. Although use of marginal donors (e.g. older, or having diabetes, cardiovascular diseases or renal insufficiency) is quite challenging, several groups (significantly the Philadelphia group) have recently achieved ! $+!remarkable success with this procedure.  Use of islets from living donors has been also suggested by the Kyoto islet transplant group, and indeed one case of a living donor has been reported [120, 121]. Although there have been promising outcomes, isolating islets from living donors is accompanied by significant challenges and risks including surgery-induced diabetes, pancreatic fistula and infection in the donor. Improving the islet preservation and isolation procedures, using breathable membranes to improve oxygenation of islets while in culture, addition of insulin-like growth factor-2 (IGF-2) to the culture [122] or co-encapsulation of islets with bioengineered IGF-2 producing cells [123], as well as numerous other approaches to improve islet cell survival have all been shown to improve the outcome of islet isolation and give a better islet yield for transplantation. Alternative sources of insulin-producing cells have generated much interest in recent years. Xeno-transplantation using neonatal porcine islets could be one approach to solve the issue of shortage of islet donors [104]. Considering the fact that porcine insulin had been used for many years for treatment of diabetes, porcine islets should be able to effectively control blood glucose levels. However, several challenges remain to be addressed such as: establishing reliable methods of isolation; immunological burdens; and risk of viral infections, importantly porcine endogenous retrovirus [104]. Neonatal porcine islets in particular are robust, resistant to hypoxia, and have been shown to maintain blood sugar control in immunosuppressed, diabetic non-human primates [124]. Another strategy is generating insulin-producing cells from glucose-responsive cells by genetic reprogramming of differentiated cells. Cheung et al., in 2000 generated glucose-responsive insulin-secretory cells from intestinal K cells [125]. Furthermore, insulin-producing cells have been generated from transdifferentiation of hepatic, bile duct epithelial and acinar ! $"!cells with encouraging results [126]. Generation of islet-like tissue from embryonic or induced pluripotent stem cells is another interesting and promising tactic [104]; using a recipient’s own stem cells would rule out the risk of an alloimmune response, although transplanting autologous cells would likely increase the risk of recurrent autoimmunity [104]. Induction of ! cell neogenesis from pancreatic progenitor/stem cells is also being investigated. Islet neogenesis-associated protein (INGAP) had shown some promise for stimulation of ! cell neogenesis from pancreatic ductal cells in a hamster model [127] and was in clinical trial [7, 128]; however, the outcome was not very positive the latest phase II clinical trial is currently suspended [129, 130]. Furthermore, using ultrasound-targeted microbubble destruction for gene delivery, Chen et al., demonstrated that delivery of betacellulin and pancreatic duodenal homeobox-1 genes could reverse STZ-induced diabetes in rats by induction of !-cell neogenesis [131], although whether such approaches can be translated to human diabetes remains to be seen. Exendin-4, a glucagon-like peptide-1 (GLP-1) receptor agonist, has been shown to improve glycemic control by improving !-cell function, increase !-cell neogenesis and proliferation and reduce apoptosis in a rat model of diabetes [132]. Recently it has been revealed that betatrophin, a secreted protein expressed in liver and adipose tissue, stimulates !-cell proliferation and may expand murine !-cell mass. The mechanism underlying this effect remains to be elucidated [133].  1.2.6.2 Improving islet engraftment and survival Following transplantation to the portal vein, islets remain ischemic, inflamed, and undergo apoptosis and necrosis until islet vasculature is established. In addition to careful donor and recipient selection, several strategies are being developed to reduce inflammation, promote revascularization, and improve transplantation outcome.  ! $#!A recent clinical report suggested that infusion of insulin and heparin pretransplantation to decrease IBMIR (heparin) and to provide !-cell rest (insulin) in the early post-transplant period, can raise the single-donor islet transplant success rate by 30% [134]. This intriguing finding needs confirmation in other clinical trials. Moreover, blockade of TNF$, an important role player in acute inflammatory destruction of ! cells, using etanercept has been suggested to reduce inflammation-mediated islet destruction. Recently, anakinra has been used for blockade of IL-1R and reduction of islet inflammation. Shapiro et al., investigated the application of anti-TNF$ antibody and IL-R blockade together using etanercept and the IL-1R anatagonist , anakinra, and demonstrated a synergistic beneficial effect of this combination therapy in promotion of islet engraftment [95].  As mentioned before, the process of isolation subjects islets to a number of mechanical and metabolic stresses. A brief culture period post isolation provides the islets with a chance to rest and recover and may allow depletion of passenger leukocytes to some extent [135].  It is suggested that several factors including ER stress, reactive oxygen species, and cytokines might trigger apoptosis in ! cells, especially post transplantation. Recently, some strategies have been investigated to prevent death of transplanted ! cells including but not limited to: (1) using cobaltic protophyrin IX [136] to activate the hemeoxygenase-1 (the inducible cytoprotective protein); (2) inducing generation of cytoprotective cyclic GMP and its dependent kinases by pretreating islets with carbon monoxide [137]; (3) inhibition of intrinsic caspase-mediated apoptosis by pretreatment of islets with Bax-inhibiting pentapeptide containing microspheres [138]; (4) using a pan-caspase inhibitor, IDN-6556, which is currently under preclinical trial [95, 138]; and (5) modification of human islets with vectors targeting X-linked inhibitor of apoptosis to inhibit apoptotic pathways [139]. Promising results in reducing islet cell ! $$!apoptosis and improving islet metabolic function have been achieved using the short-acting GLP-1 analogue, exenatide. However, the required dose of exenatide for maintaining a sufficient anti-apoptotic effect might not be tolerated by all patients, as it has been shown to cause nausea in 30% of recipients [7]. Also, controversy over its potential impact on exocrine pancreas has raised questions about its long-term use in patients [140, 141], although the results of these studies have been recently questioned [142]. Liraglutide, a long-acting analogue of GLP-1, has been shown to be associated with less nausea, improve early engraftment of islets in mice, and enhance human islet survival in vitro [95]. Another method to increase endogenous levels of GLP-1 is to block the GLP-1 degrading enzyme, dipeptidyl peptidase-IV (DPP-IV). The potential of sitagliptin, a DPP-IV inhibitor, in improving islet transplantation outcome is currently being explored in separate trials  [95], however it is still unknown whether the endogenous level of GLP-1 is enough to prevent islet cell apoptosis [7].   Delayed angiogenesis remains a significant challenge to maintain long-term survival of transplanted islets. It has been shown pro-angiogenic growth factors like vascular endothelial growth factor (VEGF) and hepatocyte growth factor (HGF), synergistically improve revascularization of islets post transplantation [143]. Furthermore, fibroblast growth factor (FGF)-21 treatment has been suggested to improve islet engraftment in mouse models. Culturing islets for 48 h with nerve growth factor before transplantation has also been shown to improve islet survival in vitro and in vivo [144] Pancreatic persufflation is a method of improved oxygen delivery that elevates adenosine triphosphate (ATP) and could ameliorate hypoxic stress [145].  One of the factors contributing to ! cell loss shortly after transplantation is IBMIR. Human islets highly express the tissue factor, which leads to platelet aggregation, islet inflammation and death.  Therefore, anti-IBMIR approaches would benefit islet transplantation ! $%!outcome. It has been shown that intravenous infusion of low molecular dextran sulphate not only increases the level of islet protective HGF but also is safe and associated with no increase of bleeding risk [146].  Incorporation of heparin onto the islet surface is also shown to abrogate IBMIR [147]. ! cells of islets are in close contact with islet ECM and basal membrane.  The interactions between islets and ECM can be most essential in the preservation of critical islet mass. Disruptions of such interactions may significantly affect islet function and engraftment. Several research teams have attempted to restore these interactions using various scaffolds. Generally ECM restoration has been shown [148] to improve islet survival and enhance glucose-stimulated insulin secretion [149-151]. A wide range of scaffolds have been tested for extra-hepatic islet transplantation including synthetic polymers [152, 153], various types of collagen [151-153], Matrigel [154], and small intestinal submucosa [155]. One of the main challenges with these matrices is biodegradability and susceptibility to gradual disintegration following transplantation.  1.2.6.3 Improving immunosuppressive strategies A critical prerequisite for successful islet transplantation is effective control of both autoimmune and alloimmune responses. Unfortunately, most of the available immunosuppressive medications to target allo- or autoimmune mediated graft rejection are toxic to the islets. Generally, orally administered immunosuppressive drugs achieve high concentrations in the portal circulation exposing the engrafted islets to potentially toxic levels of these components. Indeed, choosing an appropriate immunosuppressant considering risks and benefits is quite challenging. Furthermore, long-term immune suppression increases the risk of opportunistic infections and malignancy. As ! $&!briefly discussed before, aside from side effects associated with systemic immunosuppressive drugs, cyclosporine, tacrolimus and sirolimus (all calcineurin inhibitors) have been shown to inhibit engraftment of ! cells and islet revascularization [156] along with direct toxic effects on ! cells. Mycophenolate mofetile can inhibit !-cell replication and neogenesis. Corticosteroids may be toxic to transplanted islets by increasing insulin resistance as well as direct deleterious effects on β cells [112]. Other systemic immunosuppressants can also have negative effects on insulin sensitivity that is harmful to islet transplant success as it increases secretory demand on transplanted ! cells [156]. The Edmonton protocol suggests a corticosteroid-free immunosuppressive regimen, daclizumab (anti IL-2R antibody) for induction, high-dose of sirolimus in the first 3 months and low-dose tacrolimus for maintanance. Attempts to replace tacrolimus in the immunosuppressive regimen have been undertaken in order to enhance islet metabolic function, but such an approach could also increase the risk of rejection. Recently Shapiro et al., investigated the application of alemtuzumab, an anti-CD52 antibody that results in T cell depletion and achieved promising results [95]. Other agents with potentially less toxicity, such as OKT3 ala-ala antibody (for blockade of CD3 T cell receptor) [157] and rabbit anti-thymocyte globulin, TNF$ blockers Infliximab and etanercept, co-stimulation blockers belatacept, B7-H4.Ig and CTLA-4.Ig [7, 9], anti-lymphocyte function associated antigen (LFA)-1 antibody efalizumab have also been tested. Moreover, it has been shown that induction of C-C motif chemokine (CCL22) expression in islets protected islet graft from recurrent autoimmunity in a murine model [15].  Immunoisolation of islets is another approach to improve transplant outcomes. In this method, islets are surrounded in a semi-permeable microcapsule, which facilitates the exchange of oxygen, nutrients (including glucose), and insulin, while protecting the islets against the host ! $'!immune response. Research has aimed to use suitable materials that address the need for biocompatibility, permselectivity and durability. Several materials have been investigated for effective islet encapsulation in animal models including but not limited to: alginate, polycations and anions, agarose, and polyethylene glycol hydrogel polymer (for nanoencapsulation). With these advances in encapsulation technologies, there is an urgent need to design and test materials that are suitable for human islet transplantation, since current materials are not perfectly biocompatible and there has been reports of tissue reaction in the host [156]. 1.3 INDOLEAMINE 2,3-DIOXYGENASE The complex role of immune system in regulating the balance between inflammation and tolerance has been the subject of great interest for many years. A number of studies have focused recently on the importance of the immunoregulatory pathway of tryptophan metabolism and its role in modulation of peripheral tolerance. For almost a decade, the key enzyme of this pathway, indoleamine 2,3- dioxygenase (IDO), and its far-reaching immunomodulatory roles in different traits including neoplasia, pregnancy, autoimmune diseases, transplantation and infectious diseases, have been investigated. Many studies implicate IDO for induction of immune tolerance or as a host innate immune defense against pathogens [113].  1.3.1 Tryptophan breakdown and the role of IDO L-tryptophan is the least abundant of all essential amino acids in human body. This amino acid not only serves as a building block in protein biosynthesis, but also is a precursor for synthesis of serotonin and nicotinamide adenine dinucleotide (NAD+) [114]. While tryptophan hydroxylase ! $(!initiates degradation of tryptophan in the serotonin pathway, IDO1 and tryptophan 2,3- dioxygenase (TDO) are the rate-limiting enzymes that initiate catabolism of tryptophan through the kynurenine pathway (Figure 1.3)[114, 115]. IDO1 catalyses tryptophan to N-formylkynurenine, which is then rapidly degraded to form a series of metabolites named kynurenines and finally NAD+ [114, 116]. Serotonin can also further be converted to melatonin via activities of N-acetyltransferase and 5-hydroxyindole-O-methyltransferase [117]. The expression of TDO is mainly restricted to the liver, however, this enzyme has also been found in other tissues, like brain, mucous membranes and epididymis [115]. On the other hand, IDO1 is widely expressed in different cells and tissues, particularly immune-privileged sites, including epididymis [118], placenta [119, 120], mucosa of the gut (distal ileum and colon) [118], lung [121], and primary and secondary lymphoid organs [114, 116]. In addition to constitutive expression of IDO1 in the gut, lipopolysaccharides (LPS), type I IFNs (IFN$ and IFN!) and more potently, type II IFNs (IFN#) can induce IDO1 expression in variety of cells [122]. For instance, IDO1 is strongly induced by IFN# in monocytes and macrophages, DCs, trophoblasts, cultured fibroblasts [116] and pancreatic ! cells [115]. Type I and II IFNs have been reported to have equal potency in induction of IDO in plasmacytoid DCs [123]. IDO2 is another enzyme with IDO1-like activity, (also known as indoleamine 2,3-dioxygenase like- protein 1) [124], which has not only been identified in mammals but also in lower invertebrates [115]. The enzymatic activity of IDO1 is markedly higher than IDO2 [124].     ! $)!    Figure 1.3 Kynurenine pathway of tryptophan catabolism in mammalian cells. [Figure adapted from Ref. 158]. IDO catalyzes the first and rate-limiting step under transcriptional control by IFN#.   1.3.2 IDO expression IDO1 is encoded by a gene located on chromosome 8, region 8p12 in humans, named Indo or Ido1 gene [159]. The promoter of this gene contains a specific site for IFN#-responsive elements and two non-specific sites for IFN$ and IFN!-responsive elements [160]. IFN# mainly regulates 3-Hydroxykynurenine Tryptophan !"#$%&'()(*+),-$Kynurenine ."#$Kynurenic  Acid Antralinic  Acid Xanthurenic Acid Quinolinic Acid 3-Hydroxyanthralinic  Acid /012'-1*1-$)(*1&3')1,4-'),-$/012'-1*1-$)(*1&3')1,4-'),-$/012'-1*1-$5670+'&809),-$/012'-1*1),-$/012'-1*1),-$5670+'&80)137')9*1*:$):*+$$&80;-1),-$<2*1&9*1*:$):*+$$=7&,>7&'0?&,*93')1,4-'),-$NAD+ ! $*!Indo transcription through the Janus kinase/signal transducer and activator of transcription 1 (JAK/STAT1) pathway [159]. Upon binding, IFN# stimulates oligomerization of its receptor and activates JAK proteins. Activated JAKs phosphorylate tyrosine 440 of IFN# receptor subunit 1, providing a docking site for STAT1. Upon getting phosphorylated on tyrosine 701 and serine 727, STAT1 is diamerized and translocates to the nucleus [161] where it binds to interferon stimulatory response elements (ISREs) and #-activating sequences (GAS) in the IDO promoter and regulates expression of the IDO gene (Indo) [162]. Recently, it was reported that transforming growth factor ! (TGF!) is also able to activate Indo transcription through phosphatidylinositol 3-kinase (PI3K)/Akt and non-canonical NF-%B. This activity leads to transformation of CD8+ DCs from an immunogenic phenotype to a tolerogenic one [163]. CTLA-4 and LPS are also reported to induce IDO activity through NF-%B activity [164, 165]. Moreover, a recent study showed that LPS induces IDO expression in primary murine microglia, independent of IFN#-mediated pathway and through c-Jun-N-terminal kinase (JNK) signaling pathway [165]. Our recent study also revealed IDO expression in murine fibroblasts through LPS-mediated JNK pathway [166]. CpG-rich oligodeoxynucleotides are also inducers of IDO1 expression whose engagement with TLR9 results in the secretion of type I IFNs by plasmocytoid DCs (pDCs) [167]. Other inducers of Indo transcription include poly (I:C) (TLR3 ligand) [168], endogenous thymosin $1 [169] and hormones like chorionic gonadotropin and estrogens [170, 171]. 1.3.3 IDO structure, biochemical characteristics and regulation of expression IDO1 is an intracellular monomeric oxydoreductase containing a heme prosthetic group. This enzyme consists of 407 amino acids and the mature IDO1 has a molecular weight of 42-45 kDa ! %+![159]. Like IDO1, the IDO2 enzyme is a hemoprotein and its crystal structure has been identified to be 43% similar to IDO1 [162]. Using X-ray crystallography, it is revealed that IDO protein is folded into a large $-helical catalytic domain located in the C-terminal region of the protein, and a small $-helical non-catalytic domain near its N-terminus [162, 170, 172, 173], and a long loop connecting these two domains. The heme prosthetic group is positioned between the two domains [170, 173]. The existence of an inactive form of the enzyme shows that IDO1 might undergo post-translational modifications, which have not been fully defined yet. Incorporation of the heme prosthetic group, oxidation and reduction status of the cell, and nitric oxide production induced by pro-inflammatory cytokines are factors known to control IDO1 activity [164]. 1.3.4 IDO and modulation of immune response The mechanisms by which IDO modulates the immune response are still being elucidated. The immune-regulatory effects of IDO are in part due to broad suppression of cell proliferation caused by tryptophan starvation. Additionally, other studies suggest that downstream tryptophan metabolites could be directly responsible for some of the observed effects, as detailed below. 1.3.4.1 IDO and T cells Local tryptophan deprivation has been shown to provoke G1 phase-cell cycle arrest in effector T cells [159]. Interestingly, recapturing tryptophan does not guarantee the completion of the activation process and therefore, a second signal from the T cell receptor (TCR) is also required [163]. It has been shown that activation of a stress-responsive kinase named as general control nonderepressible 2 (GCN2) in T cells is involved not only in their proliferative arrest but also ! %"!induces anergy in these cells [172, 174-176]. Other consequences of tryptophan starvation include inhibition of T cell proliferation, induction of FOXP3 and generation of Tregs, down regulation of TCR-& chain CD8+, defective formation of memory T cells [177], and suppression of alloreactive T cells [178]. Interestingly, it has been revealed that the suppressive effect of IDO on proliferation of CD8+ T cells is stronger than that on CD4+ T cells [176]. In fact, the anti-proliferative property of IDO resulting from catabolism of tryptophan may explain its anti-tumor characteristics and direct antimicrobial effects on tryptophan-dependent intracellular pathogens and tumor cells during the course of disease. A second proposed model for IDO-induced immunomodulation is based on accumulation of kynurenines, the downstream metabolites of the tryptophan catabolism pathway (Figure 1.4). These metabolites are thought to induce apoptosis in thymocytes and Th1 lymphocytes. In addition to their direct toxicity to Th1 cells, tryptophan catabolites inhibit and induce tolerance in Th2 cells. Thus, it seems that IDO activity leads helper-T cells towards a Th2 profile [162, 179]. Furthermore, a recent study showed that 3-hydroxykynurenine suppressed CD4+ T cell proliferation while inducing development of Tregs, leading to better corneal allograft survival [180]. Another study showed that IDO activity and the presence of tryptophan metabolites resulted in a shift in cytokine responses of invariant NKT cells (an immunoregulatory subclass of T cells) towards a Th2 profile [181]. 1.3.4.2 IDO and antigen presenting cells  Tolerogenic DCs are known as the main source of IDO in the body. IDO is strongly induced by IFN$ and IFN# in plasmacytoid DCs. A combination of these two cytokines in suboptimal concentrations also induces IDO1 expression in these cells, due to observed synergic or additive ! %#!effects of these cytokines [182]. While expressed by tolerogenic DCs in response to type I/II IFNs, IDO regulates expression of type I IFNs (mainly IFN$) from CD19+ DCs as a positive feedback. Such effect leads to the stimulation of IDO expression from DCs [183].  In addition to these effects, there is evidence that 3-hydroxyanthranilic acid, a downstream metabolite of tryptophan catabolism, induces cell surface expression and secretion of HLA-G in DCs. This compound has also been shown to increase HLA-G surface expression in macrophages, but not in monocytes [182]. Therefore, one can suggest that IDO and HLA-G are two molecules involved in immunosuppressive properties of tolerogenic DCs [182, 183]. Furthermore, a separate study by Sekkai et al, [184] revealed that 3-hydroxyanthralinic acid not only inhibits the induction of nitric oxide synthase in macrophages, but also decreases the activity of this enzyme in those cells. Since nitric oxide has been identified as an inhibitor of IDO activity, it seems that the downstream metabolites of kynurenine pathway reinforce IDO function by decreasing the generation of nitric oxide [184, 185]. It has previously been shown that IFN# is able to induce Indo transcription both in immunogenic CD8' and tolerogenic CD8+ DCs. However, posttranslational inactivation of IDO in CD8' DCs interferes with the action of enzymes consequent to IDO, which results in nontolerogenic properties of these DCs [185]. It seems that addition of quinolinic acid, a downstream metabolite of tryptophan catabolism, reactivates the pathway and converts nontolerogenic CD8– DCs to suppressive CD8+DCs following treatment with IFN# [185]. 1.3.4.3 IDO and T regulatory cells Mature pDCs are able to induce IL-10-producing Tregs, which can subsequently modulate immune responses. Therefore, it seems likely that IDO contributes to peripheral expansion of ! %$!Tregs [186]. Moreover, it is well documented that Tregs expressing CTLA-4 can provoke IFN# release and subsequently high levels of IDO expression and activity by tolerogenic B7-expressing DCs [164]. This could be one of the mechanisms by which CTLA-4+ Tregs modulate immune responses. 1.3.4.4 IDO and B cells Although many studies have confirmed the suppressive effects of IDO1 (mainly on T cell survival and proliferation), some studies have suggested a more complex role for IDO1. A study by Scott et al., 2009 showed that blockage of IDO1 activity using 1-methyl tryptophan (1-MT) ameliorates B cell-mediated autoimmune responses in rheumatoid arthritis [187, 188]. In contrast, using an experimental model of autoimmune myasthenia gravis, it was shown that IFN#-treated DCs not only decreased the number of plasma cells, but also suppressed the function of B cells [189]. Moreover, it was shown that treatment with tryptophan metabolites like kynurenine, 3-hydroxykynurenine, and 3-hydroxyanthranilic acid induces B cell death [190].   ! %%!  Figure 1.4 Proposed mechanism of IDO-mediated tolerance involving DCs and regulatory T cells. The interaction of CTLA-4 molecules on T cells with B7 receptor molecules on DCs causes DCs to release IFN#, which transcriptionally activates IDO expression in an autocrine and paracrine manner in DCs. Local tryptophan depletion and the presence of proapoptotic kynurenines result in decreased clonal expansion and increased T cell apoptosis [158].   1.3.4.5 IDO and NK cells Research into the effect of IDO on NK cells has demonstrated that tryptophan deprivation and its metabolites impair the lysing activity of NK cells [190, 191]. In addition, the downstream products of tryptophan metabolism induce NK cell death [190]. Interestingly, it has been revealed that IDO expression results in downregulation of NK cell activation receptor (NKG2D) B7 T reg IDO IFN!"Dendritic Cell CTLA-4 T cells Inhibition of  Proliferation Apoptosis  Tryptophan Kynurenines ! %&!via the JNK pathway. This has been suggested as a mechanism underlying Epstein-Barr virus escape from NKG2D-mediated immune attack, since this virus is able to induce IDO expression in B cells [192].  1.3.5 Molecular mechanism underlying immunomodulatory effects of IDO  The molecular mechanism underlying immunoregulatory effects of IDO is not yet well understood. Since IDO catabolizes tryptophan, there is a possibility that it affects the pathways responding to amino acid metabolism. Several studies have shown the profound role of GCN2 kinase pathway, an amino acid sensitive pathway in IDO-mediated responses. Tryptophan deficiency, resulting from IDO activity, causes an increase in the amount of the uncharged form of transfer RNA (tRNA) [193]. Uncharged tRNA binds to the regulatory domain of GCN2 kinase leading to activation of this enzyme and triggering of the downstream signaling pathway [194]. Known as an integrated stress-response, GCN2 kinase activation can initiate cell-cycle arrest, apoptosis, differentiation and compensatory adaptation[193]. These effects vary depending on the cell type and the nature of the triggering stress [175]. Anergy is an important GCN2 kinase-mediated response that extends the immunomodulatory effects of IDO. Studies have shown that IDO-mediated anergy is antigen-specific and IDO/GCN2 pathway only affects T cells that are concomitantly activated with their specific antigen [175].  The enzymatic reactions in the kynurenine pathway start with conversion of tryptophan (by TDO or IDO) to N-formylkynurenine, that is subsequently degraded to kynurenine . In addition to kynurenine there are two other metabolic intermediates within this pathway: 3-hydroxyanthranilate and quinolinate. It has been shown that the level of the major metabolites of kynurenine pathway increases in the blood and body fluids following immune stimulation [195]. ! %'!Although tryptophan breakdown to kynurenine is a common response to IFN# treatment in many cell types [193], it seems that the conversion of 3-hydroxyanthranilate to quinolinate is only seen in hepatocytes and some immune cells [195]. Based on the finding of highly specific localization of quinolinate in immune cells, it was suggested that this metabolite plays an immunomodulatory role [195]. As described before, kynurenine, 3-hydroxykynurenine and 3-hydroxyanthranilate were shown to have additive inhibitory effects on T cell proliferation and induce apoptotic cell death [190]. Furthermore, quinolinate has been also found to induce apoptosis in a human leukemia cell line (HL-60 cells) through a caspase-mediated mechanism [196]. Morita et al. [197] have reported that 3-hydroxyanthranilate induces apoptotic cell death in monocytes via overproduction of hydrogen peroxide. Considering the data from these studies, induction of apoptosis is apparently the main mechanism underlying the inhibitory effects of kynurenine pathway metabolites on immune cells.  1.3.6 IDO and transplantation The role of IDO in inducing tolerogenic mechanisms after transplantation was first raised in a study by Miki et al. in 2001 [188], in which orthotopic murine liver allografts, normally tolerated without using immunosuppressive drugs, were rejected after recipient mice received 1-MT treatment [188]. Subsequently, overexpression of IDO by genetic modifications, using different vectors, resulted in prolonged survival of cardiac [198], lung [199], corneal [200] and islet allografts [201]. Our published data provide substantial evidence supporting the immunosuppressive roles of IDO expressed by bystander fibroblasts following either gene transduction or IFN# treatment [202-204]. Using co-culture systems we showed that IDO expression by genetically modified fibroblasts selectively suppressed the activity of bystander ! %(!human peripheral blood mononuclear cells (PBMCs), CD4+ T cells, Jurkat cells, and TPH-1 monocytes. In addition we have shown that immune but not primary skin cells or pancreatic islets are sensitive to the IDO-induced low tryptophan environment [202, 204-207][162, 164-168]. As mentioned before, the activation of GCN2 pathway, has been identified as a potential mechanism responsible for the IDO-induced suppressive effect on T cells. In our study we showed that the GCN2 signalling pathway is not activated in mouse islets in response to IDO exposure while it is strongly activated in mouse lymphocytes [208, 209], which may explain the selective responsiveness of mouse lymphocytes to IDO exposure. In a recent publication [210], we used a novel approach to study the immunosuppressive effect of IDO in islet transplantation. Bystander syngeneic fibroblasts were used as the IDO gene transfer target (using an adenoviral vector) instead of islets, to reduce the risk of cytotoxicity and loss of islet function [211]. Islets were embedded within a collagen matrix, which improves islet viability and function [212]. The findings of this study showed that local IDO expression prevents cellular and humoral alloimmune responses against islet and significantly prolongs islet allograft survival.  1.3.7 IDO and autoimmunity Loss of peripheral or central tolerance against self-antigens leads to uncontrolled immune system activation and destruction of cells and tissues, causing chronic and debilitating diseases. The activation of CD4+ T cells by autoantigens is a common process seen in autoimmune disorders. Regarding the immunomodulatory roles of IDO, this enzyme may be involved in maintaining immunological self-tolerance. Shinomia et al. provided evidence for a role for IDO in inducing self-tolerance. They showed that IFN#-stimulated DCs transferred into NOD mice can induce a long-lasting insulin independence in the recipients [213]. This might have resulted from the ! %)!immunoregulatory effects of IDO expressed in response to IFN# stimulation in DCs. This idea is further supported by the finding that DCs from NOD mice, which are prone to developing autoimmune diabetes, have a defect in IDO transcription making them nonresponsive to the tolerogenic effects of IFN# [214]. In another study, it has been shown that IDO inhibition with 1-MT accelerates the progression of type 1 diabetes [215]. The role of IDO in controlling autoimmunity has been further studied in other autoimmune diseases. Data obtained from a murine model of experimental autoimmune encephalomyelitis confirmed the immunosuppressive effect of IDO in different of stages of this disease as well. In that study, Platten et al. showed that inhibiting IDO activity by 1-MT aggravates tissue damage, inflammation and accelerates the progression of the disease [216]. 1.4 HYPOTHESIS AND SPECIFIC AIMS Despite decades of research, our knowledge about type 1 diabetes is still limited. In order to develop proper preventive or treatment strategies, it is important to more deeply understand the pathogenesis of this disease. In this study, we first aimed to understand a mechanism of impaired immunotolerance in an animal model of type 1 diabetes. Next, in order to improve islet transplantation outcome as a treatment option for this disease, we generated and applied a novel bioengineered matrix in a murine model of insulin-dependent diabetes.  NOD mice, the best available animal model of type 1 diabetes, naturally develop a T cell-mediated destruction of pancreatic ! cells, which leads to the disease. As mentioned before, it has been shown that impaired IFN#-induced IDO expression in DCs of early prediabetic female NOD mice could contribute to their defective self-tolerance[214]. Since fibroblasts are i) not professional antigen-presenting cells (unlike DCs and macrophages); ii) important residents of ! %*!ECM; and iii) play an important role in immunomodulation of islet microenvironment, under the specific aim I, we asked the question whether there exists a similar defect in IFN#-induced IDO expression in NOD dermal fibroblasts. Besides immunological hurdles of autoimmune diabetes, there are other obstacles facing treatment of type 1 diabetes.  Islet transplantation, a promising strategy to restore efficient insulin regulation in type 1 diabetes patients, has been challenged with several shortcomings. The shortage of islet donors, poor islet survival, and toxicity of immunosuppressants whose use is inevitable to prevent islet alloimmune and autoimmune rejection, limit the clinical utility of islet transplantation. Our focus in the specific aims II and III was to find a solution for these problems. We previously introduced a novel composite scaffold for islets, i.e. a fibroblast populated collagen matrix (FPCM), using which we were able to remarkably improve islet function and survival, and reduce the critical islet mass required for transplantation [217]. The FPCM not only provides a favorable platform for modulating the islet microenvironment, but also incorporated fibroblasts support and maintains collagen matrix integrity while providing islet essential growth factors [217, 218]. However, this composite was prone to gradual biodegradation and contraction jeopardizing islet graft survival over time. As such, under the specific aim II, we developed a novel bioengineered cross-linked collagen matrix (CCM) to provide optimal matrix biomimetic properties and evaluated islets function and survival when embedded within this matrix. Further, to overcome the issue of systemic immunosuppression and its side effects, we investigated the use of IDO to generate a local tolerogenic environment, through which the transplanted islets could remain viable and protected without compromising systemic immunity. Our research group has previously shown that in an IDO-mediated tryptophan deficient ! &+!microenvironment, infiltrated immune cells are not able to survive, proliferate or destroy the engrafted islets, whereas the islets themselves are unaffected by tryptophan deficiency [209]. We had previously investigated the application of genetically modified IDO-expressing fibroblasts (transduced using adenoviral vector) embedded within collagen matrix in prevention of islet graft immune rejection in STZ-induced diabetic mice [210]. In order to prolong graft survival even further, under the specific aim II, in addition to using a novel matrix, we applied and tested a more stable lentiviral IDO transduction method. Finally, under specific aim III, we tested the composite islet graft developed under specific aim II in an in vivo model. In this model, the non-rejectable islet allograft composed of lentiviral vector transduced-IDO-expressing fibroblasts and allogeneic islets embedded within CCM, was used for treatment of diabetes in STZ-induced diabetic mice.  In summary, considering the important role of fibroblasts in modulation of immune response, in this study, we firstly evaluated the IDO expression in NOD dermal fibroblasts. Secondly, we hypothesized that restoring islet ECM using our novel bioengineered matrix along with generating a long-lasting local tolerogenic environment using IDO gene transduction in co-transplanted fibroblasts would improve the islet transplantation outcome in STZ-induced diabetic mice. We evaluated our hypothesis under 3 specific aims: Aim I: To evaluate IFN#-mediated tryptophan catabolism in NOD dermal fibroblasts. Aim II: To evaluate the effect of a novel CCM populated with stable IDO producing fibroblasts on islet function and survival. Aim III: To evaluate the function/survival of an allograft consisting of allogeneic islets and stable IDO producing fibroblasts embedded within a novel CCM in STZ-induced diabetic mice.  ! &"!         Chapter 2:  MATERIALS AND METHODS              ! &#!2.1 ETHICS STATEMENT These studies have been approved by the University of British Columbia Animal Care Committee. All animals used in these studies were maintained and underwent procedures in accordance with the principles of laboratory animal care and the guidelines of the University of British Columbia Animal Care Committee. 2.2 ANIMALS Animals were housed in the Jack Bell Research Centre or Blusson Spinal Cord Centre (iCORD). Male and female NOD, C57BL/6 and Balb/c mice were purchased from the Jackson Laboratories, Bar Harbor, ME and maintained in an environmentally controlled room (temperature, 24±2°C; humidity, 55±5%, 12-h light/dark cycle) with access to food and water for at least 2 weeks. Blood glucose was monitored in NOD mice weekly using an “Accu-Chek Compact Plus” monitoring system from 8 weeks of age. The time of diabetes onset was determined as two consecutive measurements of blood glucose above 20 mM. Islet transplant recipients (C57BL/6 or Balb/c mice) were rendered diabetic by a single intraperitoneal injection of 200 mg/kg STZ (Sigma St. Louis, MO), a widely used drug for experimental induction of diabetes that is specifically toxic to beta cells.  Blood glucose levels were measured twice a week starting 3 days after STZ injection. Diabetes was defined as a minimum of two consecutive blood glucose levels "20 mM [219] and animals received islet transplants within 7 days of diabetes onset. In the meantime, mice received daily injections of NPH and regular insulin to maintain normoglycemia until receiving the islet graft.   ! &$!2.3 MOUSE DERMAL FIBROBLAST ISOLATION, CULTURE AND TREATMENTS  Dermal fibroblasts were explanted from skin punch biopsies obtained from male C57BL/6 at 8 or 12 weeks of age, diabetic female NOD mice at 12 weeks of age, diabetic male NOD mice at 24 weeks of age, and male/female prediabetic NOD mice at 8 weeks of age. Skin pieces were shaved and washed five times in sterile Dulbecco’s Modified Eagle Medium (DMEM) (Invitrogen Life Technologies, Burlington, ON, Canada), supplemented with antibiotic-antimycotic preparation (100 U/ml penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B) (Gibco, Invitrogen Incorporation, NY, USA). Cultures of fibroblasts were established as previously described [219]. Upon reaching confluence, the cells were released using trypsin, split for subculture at a ratio of 1:4, reseeded onto 75 cm2 cell culture flasks (BD Biosciences, MA). Cells were cultured in DMEM supplemented by 10% fetal bovine serum (FBS) (Hyclone Laboratories, Inc, Logan, UT, USA) and antibiotic-antimycotic preparation, and incubated in a humidified incubator at 37°C in an atmosphere of 5% CO2. Cells at passage three to five were used for this study [179].   For one part of this study, 100,000 fibroblasts per well were seeded on 6-well flat bottom cell culture plates and treated with IFN# (Sigma Chemicals, Oakville, ON, Canada) at different concentrations (0, 250, 500 or 1000 U/ml of DMEM plus 2% FBS) for either 24 (for mRNA analysis on cell pellets) or 48 h (for protein analysis on cell pellets). Conditioned media of the cells were collected for kynurenine assay 48 h post treatment. Cell pellets were obtained for analyses of IDO, MHC I and type I collagen expression.  To study IFN#-induced-STAT1 phosphorylation, cells were starved overnight and then treated with 1000 U/ml IFN-# or left untreated in DMEM plus 2% FBS. Fibroblasts were harvested at different time points (15, 30 and 60 minutes after treatment) and cell pellets were ! &%!collected for further analysis.   To study LPS-induced IDO expression, fibroblasts were seeded on 6-well flat bottom cell culture plates and treated with 1 µg/ml of LPS from Pseudomonas aeruginosa (Sigma) or left untreated in DMEM plus 2% FBS. After 24 h, fibroblasts were harvested and subjected to mRNA isolation for IDO expression analysis. To study LPS-induced IDO expression via the JNK pathway, fibroblasts were treated with 1 mg/ml of LPS in the presence or absence of SP600125 (10µM) (Sigma, St. Louis, MO), a JNK inhibitor or left untreated in DMEM plus 2% FBS for 48 h. Subsequently, cell pellets were analyzed for IDO protein expression and conditioned media were collected for kynurenine assay.  2.4 MOUSE PANCREATIC ISLET ISOLATION Pancreatic islets were obtained from 6- to 8-week-old male C57BL/6 or Balb/c mice based on a previously established method [179]. Briefly, mice were euthanized and quickly after, pancreases were distended through the pancreatic duct with 2.5 ml of Hanks’ balanced salt solution (HBSS; Life Technologies, Gaithersburg, MD) containing 1 mg/ml of collagenase (Type V; Sigma Chemical Co., St. Louis, MO). The distended pancreases were then removed and incubated at 37 °C for 10 min. Islets were purified by passing the digested pancreatic tissue through 100 µm cell strainers (BD Biosciences, Bedford, MA). Next, islets were handpicked and cultured in HAM’s F10 medium, containing 6.1 mM glucose (Sigma Chemical Co.) supplemented with 12 mM HEPES, 2 mM L-glutamine, 10% heat-inactivated fetal calf serum, 100 U/ml penicillin, and 100mg/ml streptomycin in 95% air, 5% CO2 at 37° C.   ! &&!2.5 MOUSE SPLENOCYTE AND SPLENIC DENDRITIC CELL ISOLATION Splenocytes were isolated from spleens of islet allograft rejected mice or diabetic female NOD mice. Spleen tissues were squeezed between rough edges of glass slides and suspended in RPMI-1640 medium (Hyclone Laboratories, Inc, Logan, UT, USA) supplemented by 10% FBS (Hyclone Laboratories), 100 U/ml penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B (Gibco). The cell suspension was passed through 40 µm cell strainers (BD Biosciences), washed and resuspended in PBS. The suspension was further treated with 0.155 M NH4Cl (Sigma) + 10 mM KHCO3 (Fisher) + 10 (M Na2EDTA (Fisher) for 2 min to lyse red blood cells, then washed with PBS and centrifuged; cell pellets were resuspended and cultured in supplemented RPMI-1640 until required.   Splenic DCs were isolated from diabetic female NOD mice (>12 weeks of age) and fractioned based on CD11c expression using pluriBead® cell separation kit (pluriSelect GmbH, Leipzig, Germany) according to manufacturer’s instructions. Isolated DCs were cultured within supplemented RPMI-1640 in 6-well flat bottom culture plates (Corning Incorporated, Corning, NY, USA). Viable cells were counted using a hemacytometer (Baxter Scientific) and Trypan Blue exclusion dye (Stemcell). 2.6 GENERATION OF VIRAL VECTORS Recombinant adenoviral vector carrying human IDO and green florescent protein (GFP) was constructed and used to induce IDO expression in cells [202]. Briefly, following amplification of the human IDO gene by reverse transcriptase polymerase chain reaction (RT-PCR), it was subcloned into a shuttle vector containing a green fluorescent protein gene according to ! &'!manufacturer's instructions (Q-Biogene, Carlsbad, CA, USA). The cloned plasmid was then homologously recombined with adenoviral plasmids in E. coli, BJ5183, by electroporation. The success of IDO insertion into adenoviral plasmid was confirmed by restriction endonuclease digestion. Plasmid cDNA was amplified in competent DH5α bacteria and purified in CsCl gradient in an ultracentrifuge. Adenoviral vectors carrying either GFP alone (Ade-Vect) or GFP plus human IDO gene (Ade-IDO) were then linearized by PacI digestion and used to transduce 293A package cells using Fugene-6 transduction reagent (Roche Applied Science, Laval, QC, Canada). Transduced cells were monitored for GFP expression and after three cycles of freezing in ethanol/dry ice bath and rapid thawing at 37 °C, the cell lysates were used to amplify viral particles. The viral titre and multiplicity of infection (MOI) were determined in a 96-well plate according to the manufacturer's instructions.  A lentiviral construct for expressing the IDO gene was generated using the pLC-E vector modified from the lentiviral backbone FUGWBW [220]. The IDO gene is expressed under the control of EF1-$ promoter. For selecting IDO+ cells, a sequence encoding the Blasticidine-resistance gene under the control of the ubiquitin promoter was incorporated into the vector. The human IDO gene (NM_002164; a generous gift from Dr. JM Carlin of Miami University) was generated by PCR using a full-length cDNA encoding the gene as template and the forward primer (5'- GGGGACAAGTTTGTACAAAAAAGCAGGCTTCACCATGGCACAC GCTATGGAAAACTCCTGG-3') and reverse primer (5'-GGGGACCACTTTGTACAAGAAAG CTGGGTCCTAACCTTCCTTCAAAAGGGATTTCTC-3'). The amplified PCR product was first inserted into an entry vector (pDON201) and then Gateway (Invitrogen) cloned into a lentiviral pLC-E expression vector. The plasmid was amplified in competent DH10-B bacteria and purified using the Qiagen Plasmid DNA Maxi-prep kit (Qiagen). Sequence of the IDO/pLC-! &(!E construct was confirmed by DNA sequencing analysis. Lentiviral vector carrying human IDO gene plus Blasticidin-resistance gene (Lenti-IDO) and Lentiviral vector carrying Blasticidin-resistance gene alone (Lenti-Vect) were used for transduction of fibroblasts.  2.7 TRANSDUCTION OF IDO GENE IN FIBROBLASTS  Mouse dermal fibroblasts were seeded on 6-well flat bottom cell culture plates. Fibroblasts were transduced with either Ade-IDO or the mock vector (Ade-Vect) for 72 h at an MOI of 100, or left untreated as control. Free viral particles were washed out after 30 h. Cells were washed with PBS 24 h after transduction and fresh media was added. Successful IDO transduction was confirmed by: i) monitoring GFP expression under fluorescence microscopy (Nikon, Melville, NY, HB-1010 AF); ii) detecting IDO mRNA expression using PCR; iii) detecting IDO protein expression using western blotting; and 4) measuring the L-kynurenine level in conditioned media of the transduced cells as described below.  For transduction of fibroblasts using lentiviral vector, cells were transduced with a medium containing 50% vector and 50% supplemented DMEM. One day after, the conditioned media of the cells was removed and fresh medium added. Treatment of fibroblasts with Blasticidin (8 µg/ml) on days 5 and 9 post transduction resulted in pure IDO producing cells by day 12. IDO producing cells were subcultured after and used for experimentation. Successful IDO gene transduction of cells was confirmed by measurement of L-kynurenine levels in conditioned media of the cells and PCR analysis for detection of IDO mRNA expression.   ! &)!2.8 KYNURENINE ASSAY The levels of tryptophan degradation product, L-kynurenine, were measured in conditioned media of the cells as an indicator of IDO activity, according to a previously established method [219]. Briefly, proteins were precipitated in conditioned media samples using 30% trichloroacetic acid. Samples were centrifuged and afterward, 500 µl of supernatant from each sample was incubated with one volume of Ehrlich’s reagent (Sigma Chemicals, Oakville, ON, Canada) for 10 minutes. The absorbance of the solution was measured at 490 nm using spectrophotometry.  Kynurenine levels were calculated using an equation obtained from a standard curve with defined kynurenine concentrations (0, 2, 5, 10 and 20 µg).  2.9 WESTERN BLOT ANALYSES For evaluation of IDO and type I collagen expression, harvested cells were lysed in cell lysis buffer (containing 50 mM Tris-HCL, pH 7.4; 10 mM EDTA; 5 mM EGTA; 0.5% Nonidet P-40; 1% Triton X-100 and protease inhibitor cocktail) (Sigma). Extracts were centrifuged at 14,000 RPM for 10 minutes. The total protein contents of supernatants were determined by Bradford assay [221]. A total of 100 µg of protein per sample was run on SDS-PAGE. After resolution on SDS-PAGE, proteins were transferred to PVDF membrane (Milipore Corp., Bedford, MA, USA). Membranes were immunoblotted with polyclonal rabbit anti-human IDO antibody (1:1000; Washington Biotechnology Inc, Baltimore, MD, USA) to detect IDO or monoclonal mouse anti-mouse type I collagen antibody for collagen detection.   To evaluate phosphorylation of STAT1, cell pellets were lysed in cell lysis buffer containing phosphatase inhibitor cocktail (Sigma). Samples were run on SDS-PAGE and ! &*!transferred to nitrocellulose membrane (Invitrogen, Carlsbad, CA, USA). Immunoblotting was performed using rabbit anti-mouse STAT1 antibody or rabbit anti-mouse phospho-STAT1 antibody (1:1000; Cell Signaling Technology, Inc. Danvers, MA, USA) according to manufacturer’s protocol.    Horseradish peroxidase conjugated goat anti-rabbit IgG and rabbit anti-mouse IgG were used as secondary antibodies for the enhanced chemiluminescence detection system (ECL; Amersham Biosciences, UK). 2.10 REVERSE-TRANSCRIPTASE ANALYSES Total RNA was extracted at indicated time point from fibroblasts or composite grafts using RNeasy kit (Qiagen, Maryland, USA) or Trizole® reagent (Invitrogen, Carlsbad, CA, USA) based on manufacturers’ instructions. Subsequently, cDNA was synthesized using SuperScript RT-PCR system (Invitrogen, Carlsbad, CA, USA). The primers used for detection of mouse IDO, human IDO, C57BL/6 mouse MHC I (H-2b2), NOD mouse MHC I (H-2kd), mouse type I collagen (Col 1a1), mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and mouse !-actin are listed in Table 2.1.  GAPDH and !-actin mRNA levels were used as internal controls. RT-PCR cycles were optimized for each set of primers. An annealing temperature of 55°C was chosen. The numbers of cycles were as follows: 42 cycles for mouse or human IDO, 38 cycles for type I collagen and 38 cycles for MHC I. Amplified RT-PCR products were separated by electrophoresis on a 1% agarose gel.   ! '+!Table 2.1 Primers used for RT-PCR analyses.  Gene Forward Primer Reverse Primer Product Size (bp) Mouse IDO GGCACACGCTATGGAAAACT CGGACATCTCCATGACCTTT 296 Human IDO CGGACATCTCCATG GGCACACGCTATGG 296 H-2b2 GCGAGGGTGGCTCTCACACG TCAGGGTGAGGGGCTCAGGC 554 H-2kd GGGCGGCTCTCACACGTTCC TCCCCTGCAGGCCTGGTCTC 327 Col 1a1 GACGCCATCAAGGTCTACTG ACGGGAATCCATCGGTCA 154 GAPDH TGGCACAGTCAAGGCTGAGA CTTCTGAGTGGCAGTGATGG 384 !-actin GTGGGCCGCCCTAGGCACCA CTCTTTGATGTCACGCACGA 520  Col 1a1, mouse type I collagen; H-2b2, C57BL/6 mouse MHC I; H-2Kd, NOD mouse MHC I; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.   2.11 PREPARATION OF 3D COLLAGEN AND CROSS-LINKED COLLAGEN MATRICES Collagen matrix (CM) and cross-linked collagen matrix (CCM) were prepared in a total volume of 100 µl for each treatment group. Briefly, to prepare the CM, the following ingredients were mixed in an Eppendorf tube: 350 µl 3x HAM’s F10 medium, 26 µl 0.4 N NaOH, 440 µl DMEM, 125 µl FBS, and 870 µl of 5 mg/ml acid-extracted fetal bovine type I collagen (Sigma). The final concentration of collagen in composite grafts was 2.4 mg/ml.  In order to prepare the CCM, collagen solution was neutralized with HEPES buffer and 1 N NaOH to pH 7.0 and combined with chondroitin 6-sulfate (1:5 w/w). Cross-linked gels were prepared by cross-linking for 1 h in the dark with 0.02% (w/v) glutaraldehyde at 4°C. Remaining aldehyde groups were deactivated in the dark using a glycine wash for 1 h at 4°C. Sodium ascorbate (pH 7.0) was added to each scaffold to a final concentration of 100 µM. A mixture of 5 ! '"!% polyvinyl alcohol (PVA) (Alfa Aesar, Ward Hill, MA) in DMEM, pH 7.5 (Gibco, Invitrogen, Burlington, Canada) and 6.25 mM sodium tetraborodecahydrate, pH 8.0 (Sigma, Oakville, Canada) (final concentration 0.011 wt.% sodium tetraborohydrate, 0.2% (w/v) PVA was prepared and combined with glycine-washed cross-linked scaffolds and allowed to settle overnight at 4°C prior to use. All scaffolds were prepared to a final collagen concentration of 3 mg/ml [62]. Liquid composites were freshly prepared and stored at 4°C or on ice until used.  For in vitro studies, mouse islets and/or isolated fibroblasts were added to CM and CCM before solidification to prepare fibroblast populated CM (FPCM) and fibroblast populated CCM (FP-CCM) respectively. For preparation of the FP-CCM, either non-transduced or Lenti-IDO transduced C57BL/6 fibroblasts were used with the ratio of 100 fibroblasts per islet. Each treatment group was prepared in triplicates within 48-well plates in a total volume of 100 µl and immediately transferred to a humidified incubator at 37 ° C in an atmosphere of 5% CO2 for 1 h for polymerization. The composite grafts were maintained in 95% air, 5% CO2 at 37 ° C [62].  For another part of in vitro studies, murine islets were embedded within CCM either alone or co-cultured with non-transduced or Lenti-IDO transduced C57BL/6 fibroblasts. A group of islets were cultured in two dimensional cell culture condition as a control group. A group of islets were embedded within the CCM along with Lenti-Vect fibroblasts. Each treatment group was prepared in triplicates within 48-well plates consisting of 50 islets per well and ratio of 100 fibroblasts per islet in a total volume of 100 µl. Following preparation, mixtures were immediately transferred to a humidified incubator at 37 °C in an atmosphere of 5% CO2 for 1 h for polymerization. The composite scaffolds were maintained in incubator for up to 30 days for further analyses. For in vivo experiments, 500 islets were embedded within the matrices. After ! '#!incorporating the islets, the CM graft was immediately transferred to a humidified incubator at 37 °C in an atmosphere of 5% CO2 for 1 h for polymerization and maintained overnight until transplantation. The CCM is a liquid matrix that gets solidified in less than 15 minutes at 37 °C and is injectable using a catheter. Therefore, the CCM liquid composites were freshly prepared within a few h before transplantation and after mixing with islets and fibroblasts were kept on ice until used. A ratio of 100 fibroblasts per islet was considered for the in vivo experiments. 2.12 DETERMINATION OF ISLET VIABILITY The viability of islets cells embedded in CM or CCM or cultured in regular culture dish was determined in vitro on day 1, 15 or 30 post culture, by the Live/Dead ® Viability/Cytotoxicity Assay Kit (Invitrogen, Eugene, OR) according to manufacturer’s instructions. Two fluorescent dies are provided in this kit which determine live and dead cells by detection of intracellular esterase activity (Calcein AM die) and plasma membrane integrity (ethidium homodiamer) respectively. Cultured composites were washed twice with PBS and then 100 µl of Calcein (2 µM)/ethidium (4 µM) solution was added to each well. After 30 min of incubation in the dark at 37 °C, plates were examined under an inverted fluorescent microscope. A Zeiss Axioplan II microscope and AxioVision image analysis software were used to obtain the images. The ratio of cell death was evaluated by measuring the red fluorescent intensity normalized to the islet area using the Image J software.   2.13 ASSESSMENT OF ISLET FUNCTION In vitro function of islets was evaluated using glucose-stimulated insulin secretion assay as ! '$!follows. FPCM and FP-CCM composites were digested by incubating composites in type I collagenase solution (4 mg/ml; Sigma) at 37 ºC for 5 min. Collagenase activity was then stopped by adding supplemented HAM’s F10 medium and islets were hand-picked from digested matrices under microscope. Islets (50 islets/well, duplicates) were then pre-incubated (1 h) in Krebs-Ringer bicarbonate buffer (KRBB) containing 10 mM HEPES (pH 7.4), 0.25% BSA, and 1.67 mM glucose in 48-well plates at 37°C followed by 1 h incubation in KRBB containing either 1.67 mM (basal) or 16.7 mM glucose (glucose stimulated) insulin release. Islets in each well were lysed in 100 µl lysis buffer containing 1 M HCl and 0.1% BSA. Incubation media and islet lysates were collected, centrifuged (12,000 rpm, 10 min, 4°C), and the supernatants were frozen at -20°C until assayed. Insulin levels in media and islet lysates were measured using a mouse insulin ELISA kit (ALPCO, Salem, NH). 2.14 TRANSPLANTATION OF ISLET-FIBROBLAST COMPOSITE SCAFFOLDS C57BL/6 or Balb/c male mice at 6-8 weeks of age were rendered diabetic by a single intraperitoneal injection of STZ as described before. Composite three-dimensional islet grafts were prepared by embedding 500 allogeneic or syngeneic islets within FPCM or FP-CCM. Islet grafts were transplanted into the renal subcapsular space of STZ-induced diabetic mice. After transplantation, blood glucose levels were measured using an “Accu-Chek Compact Plus” monitoring system. Grafts were deemed functioning when blood glucose levels decreased to <10 mM. All animals were cared for according to the guidelines of the institutional animal policy and welfare committee.  ! '%!2.15 HISTOLOGY ANALYSES For in vitro analyses, composite scaffolds were harvested at indicated time points, and CCMs were digested using type I collagenase (1 mg/ml; Sigma) at 37°C for 20-30 min. Then, islets were hand picked and fixed in 4% paraformaldehyde. Paraffin-embedded sections were rehydrated and nonspecific binding was eliminated by incubating the tissue sections in 5% goat serum (Sigma Chemical Co.) and 2% fetal bovine serum (Sigma Chemical Co.) in PBS for 60 min. Sections were double immunostained for insulin/glucagon or insulin/cleaved caspase-3 according to our previously established method: sections were incubated overnight at 4°C with primary antibodies, washed three times with PBS for 5 min each, and then incubated with secondary antibodies for 45 min in the dark. Finally, after washing with PBS three times for 5 min each, samples were mounted in Vectashield H-1200 (Vector Laboratories, Inc., Burlingame, CA) containing 4,6-diamidino-2-phenylindole (DAPI) for nuclei staining. For in vivo studies, graft-bearing kidneys and draining lymph nodes were retrieved at different time points. Cross-sections were prepared from paraffin embedded kidneys and lymph nodes of three animals per group and at least three sections per animal. Tissues were processed into 5 µm paraffin sections. For kidney sections, three sections (separated by 100 µm) were graded and averaged. Paraffin-embedded kidney sections were stained with H&E or double-immunostained for insulin/CD3 or single stained for FOXP3 for evaluation of islet graft function, T cell and FOXP3+ immune cell infiltrations respectively. Draining lymph nodes were stained for FOXP3+ cell infiltration. Sections were counter-stained with DAPI for detection of nuclei. The primary and secondary antibodies used for immunostaining are shown in Tables 2.2 and 2.3 respectively. A Zeiss Axioplan II microscope and AxioVision image analysis software were used to obtain the images. ! '&!Table 2.2 Primary antibodies used for immunostaining.  Antibody Source Dilution Manufacturer Anti-human IDO  Rabbit 1:50 Abcam, Cambridge, MA Anti-insulin Guinea pig 1:750 Dako, Mississauga, Canada Anti-glucagon Rabbit 1:500 Dako, Mississauga, Canada Anti-cleaved caspase-3 Rabbit 1:100 R&D systems Anti-CD3 Rabbit 1:100 Abcam, Cambridge, MA Anti-FOXP3 Rat 1:50 ebioscience, California, USA  Table 2.3 Secondary antibodies used for immunostaining.  Antibody Source Fluorochrome Dilution Manufacturer Anti-guinea pig IgG Goat Rhodamine 1:2000 Abcam, Cambridge, MA Anti-guinea pig IgG Goat FITC 1:2000 Abcam, Cambridge, MA Anti-rabbit IgG Goat Rhodamine 1:2000 Chemicon International Anti-rabbit IgG Goat FITC 1:2000 Chemicon International Anti-rat IgG Rabbit Rhodamine 1:3000 Abcam, Cambridge, MA  2.16 EVALUATION OF FIBROBLASTS PROLIFERATION In order to analyze the proliferation of fibroblasts in CCM and CM, the same number of fibroblasts (25,000/well) was seeded in plastic culture plates, and CM or CCM in 6-well plates.  Cells were harvested after 1, 10, 20 and 30 days by collagenase digestion. Fibroblasts were then resuspended in 0.04% trypan blue dye and counted using a hemocytometer where cell viability was determined by the proportion of cells that excluded the dye. The fibroblast proliferation rate was measured by dividing the number of fibroblasts for the day of measurement by the number of fibroblasts on day 1; this value was reported as fibroblast count (fold of day 1). ! ''!2.17 VIABILITY AND IDO EXPRESSION OF TRANSDUCED FIBROBLASTS IN CCM Lenti-IDO or Lenti-Vect fibroblasts were embedded within CCM for 2 weeks. Fibroblasts viability was examined using Live/Dead® Viability/Cytotoxicity assay kit (Invitrogen) on day 14 post culture. Paraffin-embedded sections were immunostained for human IDO as described previously [210]. Rabbit anti-human IDO antibody and rhodamine-conjugated anti-rabbit IgG were used as primary and secondary antibodies, respectively (Tables 2.2 and 2.3).  2.18 LASTING EFFECT OF IDO EXPRESSION IN COMPOSITE SCAFFOLDS FPCM and FP-CCM were prepared firstly by embedding different numbers (50 )103, 100 )103, 150 )103, 200 )103) of either Lenti-IDO or Lenti-Vect fibroblasts within either CM or CCM. Conditioned media of the composite scaffolds were collected every 3 days for up to 90 days. L-kynurenine level was measured in the collected samples as an indicator of IDO activity, as described previously [219]. Furthermore, FPCM and FP-CCM in which 200 )103 fibroblasts were incorporated were collected every 15 days and subjected to RT-PCR assay for evaluation of IDO mRNA expression for up to 90 days. 2.19 ANTIGEN SPECIFIC PROLIFERATION ASSAY For some in vitro experiments, splenocytes isolated from Balb/c islet allograft rejected C57BL/6 mice were labeled with 5-(and 6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) according to manufacturer’s instruction (ebioscience Inc). Next, 5 ) 104 Lenti-IDO C57BL/6 fibroblasts were seeded into 6-well flat bottom plates (Corning Inc.) and co-cultured with 100 ! '(!Balb/c islets per well. Subsequently, 25 ) 104 CFSE-labeled splenocytes were added to the co-cultures in RPMI medium with 0.05% !-mercapto-ethanol (Fisher) in the presence or absence of 1-MT, a competitive inhibitor of IDO, at the final concentration of 800 µM. A group of islets were co-cultured with CFSE-labeled splenocytes and non-transduced C57BL/6 fibroblasts and considered as control group. After 72 h, flow cytometry was performed to evaluate the proliferation of primed splenocytes by allogeneic islets in the presence or absence of IDO activity. 2.20 STATISTICAL ANALYSIS Data in graphs are represented as mean ±standard error of the mean (SEM) or ±standard deviation (SD) as indicated in the figure legends. The statistical analysis was performed by one-way ANOVA followed by post hoc evaluation using Student’s t-test for multiple comparisons to ensure the proper distribution of variances. Survival curves were generated using Kaplan-Meier life-table analysis and compared using the log-rank test. P-value of less than 0.05 was considered statistically significant. For all the experiments, “n” refers to the number of independent experiments. Replicates were used in each experiment and the mean of replicates was considered as the outcome of each independent study. For each independent experiment, cells or islets were isolated from different animals.       ! ')!         Chapter 3:  MECHANISM UNDERLYING DEFECTIVE  IFN#-INDUCED IDO EXPRESSION IN  NON-OBESE DIABETIC  MOUSE FIBROBLASTS       ! '*!3.1 INTRODUCTION The importance of IDO, a rate-limiting enzyme in metabolism of tryptophan, in induction of self-tolerance and feto-maternal tolerance during pregnancy in mice has been well documented [158, 222].  It is well established that IDO plays a substantial role in modulation of T cell-mediated immune responses. IDO-induced local tryptophan depletion and accumulation of kynurenine, a tryptophan metabolite, can both contribute to the suppression of T cell-mediated immune responses [172, 175, 177, 179, 180].   Among different subtypes of immune cells, little attention has focused on the role of DCs in the pathogenesis of type 1 diabetes. Following maturation, the expression of MHC I, MHC II, and costimulatory molecules including CD40, CD80 and CD86 are upregulated on their membrane, increasing DCs ability to activate naïve T cells. Therefore, due to their elevated capacity to stimulate T cells and drive T cell differentiation, DCs would be expected to be involved in the development of ! cell-specific T cell reactivity. In the NOD mouse model of type 1 diabetes, DCs are among the first groups of immune cells infiltrating the islets of Langerhans in the early stages of disease. It has been shown that DCs from NOD mice exhibit characteristics different from other general strains, which may contribute to disease pathogenesis. Recent studies have shown that DCs derived from bone marrow or spleen of NOD mice exhibit enhanced levels of activation of NF-%B and therefore antigen presentation ability following various types of stimulation compared with DCs from C57BL/6 or BALB/c mice [223]. It is also revealed that NOD DCs have higher capacity to stimulate CD8+ T cell activation.   Recently, Grohmann and colleagues [214] have shown that tolerogenic properties of CD8+ DCs, mediated by IDO expression, is impaired in prediabetic NOD mice. They found that IFN#, ! (+!a potent inducer of IDO expression, selectively fails to induce tryptophan catabolism in CD8+ DCs of prediabetic female NOD mice. They suggested that temporary blockade of a known IFN# signaling pathway, JAK/STAT1, caused by IFN#-induced-peroxynitrite generation, explains this phenomenon. As IFN# is a potent IDO inducer in many different cell types, including fibroblasts, and considering that fibroblasts are known to act as non-professional APCs [224, 225], here we asked whether there exists a similar defect in IFN#-induced IDO expression in dermal fibroblasts from NOD mice and if so, whether this is limited to pre-diabetic female mice.  To address this, we investigated IFN#-induced-IDO expression and tryptophan catabolism in dermal fibroblasts of non-diabetic (C57BL/6), prediabetic (8 weeks of age) male and diabetic female (12 weeks of age) NOD mice.  3.2 RESULTS 3.2.1 IFN"  fails to induce tryptophan catabolism and IDO expression in NOD splenic DCs and dermal fibroblasts  To confirm the previous findings by Grohmann et al. [214], splenic DCs were isolated from diabetic female NOD (>12 weeks of age) and C57BL/6 mice and treated with 1000 U/ml of IFN# or left untreated for 24 hours. Cell pellets were subjected to RT-PCR analysis for evaluation of IDO mRNA expression (Figure 3.1A). Consistent with previous investigations, we found that IFN# failed to induce IDO mRNA expression in NOD splenic DCs, unlike DCs from Bl/6 mice. Next, we examined IDO activity and tryptophan catabolism in dermal fibroblasts from non-diabetic C57BL/6 (male, 12 weeks of age) and diabetic NOD (female, 12 weeks of age; male, 24 weeks of age) mice. Fibroblasts were treated with different doses of IFN# for 48 ! ("!hours. The conditioned media of the cells were then subjected to kynurenine measurement. The results shown in Figure 3.1B indicate a dose-dependent increase (11.4±1.8, 15.5±1.7 and 17.1±1.8 µg/ml respectively) in kynurenine levels in conditioned media of C57BL/6 cells that were treated with various concentrations of IFN#, relative to that of untreated control (2.08±1.09).  However, there was no increase in kynurenine levels produced by IFN#-treated female NOD fibroblasts (1.78±0.09, 1.69±0.09 and 1.81± 0.08 µg/ml respectively) or male NOD fibroblasts (1.81±0.09, 1.72±0.08, 1.78± 0.08 compared to that of non-treated ones (1.57±0.05 and 1.85±0.03 µg/ml for female and male NOD fibroblasts respectively); suggesting that no active IDO protein was present in NOD fibroblasts.  To determine at which level (mRNA production or protein expression) this impairment occurred, fibroblast RNA and cell lysates were subjected to RT-PCR and western blot analysis, respectively. The results showed that, similar to the enzymatic activity, IFN# failed to induce IDO expression at the protein (Figures 3.1C and 3.1D) and mRNA (Figures 3.1E and 3.1F) levels in NOD dermal fibroblasts. Figures 3.1D and 3.1F represent the quantitative analysis of the data shown in Figures 1C and 1E, respectively. Expression of IDO mRNA and protein were markedly induced in C57BL/6 cells in a dose-dependent manner in response to IFN# treatment. These data suggest that IFN#-mediated-tryptophan catabolism is impaired in dermal fibroblasts of NOD mice.        ! (#!  Figure 3.1 Effect of IFN# on IDO expression in NOD splenic DCs and dermal fibroblasts. Splenic CD11c+ DCs from non-diabetic male C57BL/6 mice, diabetic female NOD mice at 12 weeks of age, and diabetic male NOD mice at 24 weeks of age were treated with 1000 U/ml of IFN# or left untreated for 24 hours. Skin fibroblasts from indicated mice were treated with increasing doses of IFN# (0, 250, 500, 1000 U/ml) for 48 hours. A: IDO mRNA expression in C57BL/6 NOD IDO  GAPDH IFN! (U/ml)      0   1000    0    1000 CD11c-DCs IDO/GAPDH ratio 0.500 0.375 0.250 0.125 0.000  IFN! (U/ml)           0                   1000    IDO/"-actin ratio 3.00 2.25 1.50 0.75 0.00 IFN! (U/ml)        0              500            1000    IDO C57BL/6 NOD (f) NOD (m) IFN! (U/ml)      0    1000      0     1000      0     1000   GAPDH IDO "-actin C57BL/6 NOD (f) NOD (m) IFN! (U/ml)    0    500 1000     0   500 1000    0   500 1000   Kynurenine level (µg/ml) IFN! (U/ml)   0         250        500       1000    20 15 10 5 0  C57BL/6 Diabetic NOD (f) Diabetic NOD (m) A B C D E F ! ($!the splenic DCs, B: Kynurenine level in conditioned media of the cells was measured as an indicator of IDO activity. C: IDO protein expression, E: IDO mRNA expression, D and F: the mean±SEM ratio of densities of IDO to !-actin at the protein and mRNA levels, respectively. !-actin and GAPDH were used as loading controls for western blotting and RT-PCR respectively. * corresponds to significant difference between C57BL/6 and NOD fibroblasts treated with the same concentration of IFN# (p<0.001, n=6; refers to the number of independent experiments). ND: not detected, (f): female, (m): male.     To examine the effect of gender and status of diabetes on IDO induction by IFN#, dermal fibroblasts isolated from prediabetic male and female NOD mice (8 week of age) and control male C57BL/6 mice (8 weeks of age) were treated with IFN# for 48 hours.  The level of kynurenine, as an index for IDO activity, was measured and our result showed that in contrary to C57BL/6 cells, fibroblasts from neither male nor female prediabetic NOD mice were able to catabolize tryptophan to kynurenine (kynurenine levels 17.05± 0.28 µg/ml for C57BL/6 cells versus 1.68±0.22 and 1.71±0.14 µg/ml for female and male prediabetic NOD fibroblasts respectively) (Figure 3.2A). Comparing fibroblasts from female and male prediabetic NOD with control cells, we found that regardless of gender, dermal fibroblasts from 8 week old prediabetic NOD mice fail to express IDO following IFN# treatment. However, C57BL/6 cells significantly expressed active IDO (Figure 3.2B). Quantitative analysis of the blots was also consistent with this result (Figure 3.2C). ! (%! Figure 3.2 Different effect of IFN# on IDO expression in dermal fibroblasts of C57BL/6 and prediabetic NOD mice. Dermal fibroblasts from prediabetic (8 weeks of age) male and female NOD mice failed to respond to IFN# induced IDO. Dermal fibroblasts isolated from C57BL/6 male mice of 8 weeks of age as control, and aged matched male or female prediabetic NOD mice were treated with 1000 U/ml of IFN# for 48 hours. A: Kynurenine levels in conditioned of treated cells, B: IDO expression at the protein level, C: mean±SEM ratio of densities of IDO to !-actin at protein control group treated with IFN#. !-actin expression showed equal loading of proteins. *corresponds to significant difference between C57BL/6 and NOD fibroblasts treated with the same concentration of IFN# (p<0.01, n=6; refers to the number of independent experiments). ND: not detected.   C57BL/6 Diabetic NOD (f) Diabetic NOD (m) IFN!          +  +     + 2.0 1.5 1.0 0.5 0.0 IDO/"-actin ratio C Kynurenine level (µg/ml) 20 15 10 5 0 IFN!          +  +    + IFN!        +        +         + IDO "-actin A B C57BL/6 NOD(f) NOD(m) ! (&!3.2.2 IFN# induces MHC I expression in NOD dermal fibroblasts  To get insights into the mechanism underlying impaired responsiveness of NOD fibroblasts to IFN#, we first asked whether the expression of other IFN#-mediated genes is also defective in these cells. As a well-established molecule upregulated by IFN# [226, 227], we evaluated the MHC class I mRNA expression by RT-PCR in dermal fibroblasts of 12 week old C57BL/6 (male) and diabetic NOD (female) mice following treatment with IFN#. Activation of JAK/STAT1 signaling pathway has been previously shown to be involved in IFN#-mediated MHC class I upregulation [227]. In addition to this, there is compelling evidence suggesting the role of NF-#B pathway in the regulation of MHC class I gene expression [228]. Therefore, any defect in MHC class I expression due to an aberrant JAK/STAT1 pathway, could be compensated for through the activity of the NF-#B pathway. As shown in Figure 3.3A, IFN# significantly increased MHC class I mRNA expression in both C57BL/6 and NOD fibroblasts. The quantitative analysis of these signals revealed a significant difference in MHC class I expression between IFN# treated NOD fibroblasts and that of untreated control (Figure 3.3B). These data show that IFN# can bind to its receptor at the surface of NOD fibroblasts and stimulate the expression of MHC class I.      ! ('!    Figure 3.3 MHC class I mRNA expression in fibroblasts isolated from control and NOD mice. Cells were treated with 0 or 1000 U/ml IFN# for 48 hours. A: RT-PCR analysis of MHC I mRNA expression. B: the mean±SEM ratio of densities of MHC I to GAPDH. Solid and open bars represent C57BL/6 and NOD fibroblasts respectively. GAPDH was used as loading control. *denotes a significant difference between C57BL/6 and NOD fibroblasts treated with IFN# in terms of MHC I expression (p<0.05, n=5). **corresponds to significant difference between cells from the same strain treated with 0 or 1000 U/ml of IFN # (p<0.01, n=5; refers to the number of independent experiments).     MHC-I GAPDH C57BL/6 NOD IFN!         "        +          "        + IFN!         "                +   MHC-I/GAPDH ratio 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 A B C57BL/6 NOD ! ((!3.2.3 IFN# reduces collagen expression in NOD dermal fibroblasts  It is well documented that IFN# reduces the expression of collagen through the activity of CCAAT/enhancer-binding protein ! (C/EBP!), a JAK/STAT1-independent mediator [229]. To further investigate the responsiveness of other genes to IFN#, fibroblasts from 12 weeks old male C57BL/6 and diabetic female NOD fibroblasts were treated with IFN# or left untreated for 48 hours. The result showed a significant difference in type I collagen expression between untreated and treated NOD fibroblasts at the protein level (Figure 3.4A). Interestingly, quantitative analysis showed that the expression of type I collagen in untreated NOD fibroblasts was significantly lower compared to that of C57BL/6 fibroblasts (Figure 3.4B). Consistent with protein expression, RT-PCR showed that IFN# was capable of suppressing type I collagen mRNA expression significantly in both treated C57BL/6 and NOD fibroblasts compared to untreated cells (Figure 3.4C). The quantitative analysis of RT-PCR assay is shown in Figure 3.4D.         Taken together, these results suggest that IFN# receptor and other IFN#-mediated protein expression pathways seem to be functional and further that the defect is specific for IFN#-induced-IDO expression pathway in dermal fibroblasts of NOD mice.    ! ()!  Figure 3.4 Type I collagen expression in dermal fibroblasts from C57BL/6 and NOD mice.  Type I collagen expression in dermal fibroblasts from C57BL/6 (solid bars) and NOD (open bars) mice was evaluated by western blot and RT-PCR analyses. Cells were exposed to 0 or 1000 U/ml of IFN# for 48 hours before analysis. A: Type I collagen expression at the protein level. C: Type I collagen expression at mRNA level. B and D represent the mean±SEM ratio of type I collagen to !-actin at protein and mRNA levels respectively. !-actin was used as loading control in both western blotting and RT-PCR assays. *demonstrates a significant difference between C57BL/6 and NOD fibroblasts treated with IFN# in terms of type I collagen expression (n=6, p<0.05). **corresponds to a significant difference between cells from the same strain treated with 0 or 1000 U/ml of IFN# (p<0.05, n=6; refers to the number of independent experiments).    C57BL/6 NOD Collagen  !-actin IFN"       #     +      #    + C57BL/6 NOD Collagen  !-actin IFN"       #     +      #    + IFN"           #           +      Collagen/!-actin ratio Collagen/!-actin ratio 2.5 2.0 1.5 1.0 0.5 0.0  1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 IFN"           #           +      C57BL/6 NOD A B C D ! (*!3.2.4 IDO gene-transduced NOD fibroblasts express IDO To gain in-depth perspective on the mechanism underlying impaired IFN#-induced tryptophan catabolism in NOD dermal fibroblasts, we next asked whether the ribosomal protein production machinery is intact in these cells.  In other words, we investigated whether the fibroblasts are capable of translating IDO mRNA to functional IDO protein. To achieve this, we employed an adenoviral vector bearing IDO and GFP reporter genes (Ade-IDO) to transduce C57BL/6 (12 weeks of age, male mice) and diabetic NOD (12 weeks of age, female mice) dermal fibroblasts as described previously [204]. A mock vector was used as control for IDO bearing vector. A separate batch of fibroblasts from both NOD and C57BL/6 mice were also treated with 1000 U/ml of IFN#. Microscopic evaluation of transduced NOD fibroblasts after 72 hours, confirmed the viability of the cells and successful transduction as indicated by GFP expression under UV light (Figure 3.5A). To investigate the efficacy of IDO transduction in fibroblasts, conditioned media of the cells treated with IFN# or transduced with Ade-IDO were collected for kynurenine measurement. The result showed a significant increase in kynurenine levels in conditioned media of Ade-IDO transduced NOD (12.13±0.94 µg/ml) and C57BL/6 (14.09±0.10 µg/ml) fibroblasts compared to untreated cells (2.09±0.14 and 1.94±0.22 for NOD and C57BL/6 fibroblasts respectively) suggesting the functionality of expressed IDO in these cells. In contrary to NOD fibroblasts (kynurenine level 1.74±0.19 µg/ml), C57BL/6 fibroblasts were able to catabolize tryptophan following IFN# treatment markedly (kynurenine level 16.97± 0.10 µg/ml)(Figure 3.5B).     ! )+!    Figure 3.5 Tryptophan catabolism and microscopic evaluation of IDO expression in dermal fibroblasts following transduction with Ade-IDO vector. Fibroblasts were treated with either blank medium or IFN# (1000 U/ml), or transduced with Ade-IDO vector carrying GFP at an MOI of 100. A: IDO and GFP expression in Ade-IDO transduced fibroblasts. B: Increased levels of kynurenine indicate that NOD and C57BL/6 fibroblasts express active IDO following IFN# (48 h post-treatment) or Ade-IDO treatment (72 h-post transduction). *denotes a significant difference between related bars (p<0.001, n=6; refers to the number of independent experiments). Data are reported as mean±SEM. UT: untreated.     20 15 10 5 0 !"!"Kynurenine level (µg/ml) UT IFN! Ade-IDO Bright Field Ade-IDO fibroblasts A B NOD C57BL/6 ! )"! For assessing IDO expression at the protein and mRNA level, Ade-IDO transduced cells were harvested and both protein and mRNA expression were evaluated by western blot analysis and RT-PCR, respectively.  The results confirmed a high level of IDO expression in Ade-IDO transduced NOD and C57BL/6 cells relative to vector treated cells at both the protein level (Figure 3.A, B)by RT-PCR analysis of mRNA showed that Ade-IDO transduced NOD and C57BL/6 fibroblasts markedly expressed IDO mRNA (Figure 3.6C. D).  The quantitative analysis of these data is shown in Figure 6D. Collectively, these results elucidate that the unresponsiveness of NOD fibroblasts to IFN# treatment in terms of IDO expression is not due to a defect in translation of the IDO mRNA to IDO protein. As such, these findings suggest that this defect is likely to be upstream in the IFN#-induced-IDO specific signaling pathway.         !! )#!  Figure 3.6 IDO protein and mRNA expression in Ade-IDO transduced cells. Dermal fibroblasts from C57BL/6 (solid bars) and NOD (open bars) mice were transduced with Ade-IDO or mock vector. A: IDO expression was analyzed by western blot, C: IDO expression was analyzed by RT-PCR. B and D: the mean±SEM ratio of IDO to !-actin at the protein and GAPDH at mRNA level (n=6; refers to the number of independent experiments). !-actin and GAPDH were used as a loading control for protein and mRNA expression respectively. ND: not detected.     IDO !-actin Ade-IDO  +       "       +              Mock   C57BL/6 NOD NOD C57BL/6 Ade-IDO    +       "      +    Mock  C57BL/6 NOD 2.0 1.5 1.0 0.5 0.0 IDO/!-actin ratio C57BL/6 NOD IDO GAPDH Ade-IDO  "          +       "          +   IDO/GAPDH ratio 3 . 0 0 2.25 1.50 0.75 0.00 Ade-IDO       "                +   A B C D ! )$!3.2.5 Defective STAT1 phosphorylation is responsible for impaired IFN# -induced tryptophan catabolism in NOD fibroblasts  It is well established that IFN# activates phosphorylation of JAK and subsequently STAT1. The phosphorylated STAT1 then translocates to the nucleus and initiates transcription of responding genes [161, 230]. Knowing this, we monitored STAT1 phosphorylation in diabetic female NOD and male C57BL/6 dermal fibroblasts following treatment with IFN# at different time points (0, 15, 30 or 60 minutes). The result showed that following treatment, IFN# induced STAT1 phosphorylation in C57BL/6 fibroblasts showing a peak at 30 minutes. However, it failed to induce STAT1 phosphorylation in NOD fibroblasts (Figure 3.7A). The quantitative analysis of these data is shown in Figure 3.7B. The expression of phosphorylated STAT1 was normalized to the expressions of total STAT1 and !-actin. This finding indicates that a defect in STAT1 phosphorylation should, at least in part, be responsible for IFN# failure to induce IDO expression in these cells.             ! )%!   Figure 3.7 IFN#-induced-STAT1 phosphorylation in C57BL/6 and NOD dermal fibroblasts. Following starvation for 18 hours, dermal fibroblasts from NOD and C57BL/6 mice were left  untreated or treated with 1000 U IFN# per ml of DMEM plus 2% FBS for 15, 30 or 60 minutes. Cell lysates were collected for western blot analysis. A: STAT 1 phosphorylation shown by western blotting. B: the mean±SEM ratio of phospho-STAT1 (P-STAT1), to the ratio of !-actin to total STAT1. Total STAT1 and !-actin expressions were used as loading controls. *denotes a significant difference between related bars (p<0.01, n=3; refers to the number of independent experiments). UT: untreated, ND: not detected.          UT              15                30               60             IFN!(min)!P-STAT1 STAT1 "-actin UT     15      30     60        UT   15      30     60      IFN! (min)!C57BL/6 NOD 25 20 15 10 5 0 (P-STAT1 # "-actin/STAT1) ratio  "! "!A B NOD C57BL/6 ! )&!3.2.6 LPS induces IDO expression in NOD dermal fibroblasts  Using another IDO inducer, we next examined whether the expression of IDO via an IFN#-independent pathways is also defective or not. To investigate this, C57BL/6 and NOD (12 weeks of age, male and female respectively) dermal fibroblasts were treated with 1 µg/ml of LPS, derived from Pseudomonas aeruginosa [231] for 24 hours. Using primary murine microglia, it has been previously shown that LPS can induce IDO expression via an IFN#-independent mechanism with a peak of enzymatic activity at 24 hours. This mechanism depends upon activation of the JNK pathway [165].  Our result showed that LPS was able to induce IDO mRNA expression in both C57BL/6 and NOD dermal fibroblasts (n=3, p<0.05)(Figure 3.8A). Figure 3.8B represents the quantitative analysis of this data. Next, the control and test cells were treated with 1 µg/ml of LPS in the presence or absence of SP600125 (10µM), a JNK inhibitor, for 48 hours. The result of western blot analysis showed that both C57BL/6 and diabetic NOD fibroblasts markedly expressed IDO following treatment with LPS. This effect was partially reversed in the presence of the JNK inhibitor (Figure 3.8C). The quantitative analysis of the data is shown in Figure 3.8D. The result of kynurenine assay was in parallel with the protein assay and the data showed that both groups were able to significantly catabolize tryptophan, while SP600125 partially reversed this effect (Figure 3.8E).    ! )'!   Figure 3.8 LPS-induced IDO expression in C57BL/6 and NOD dermal fibroblasts. In panels A and B, C57BL/6 and NOD were treated with 0 or 1 µg of LPS from Pseudomonas aeruginosa per ml of DMEM plus 2% FBS, for 24 hours. A: RT-PCR analysis of IDO mRNA expression. B: mean±SEM ratio of densities of IDO to !-actin. GAPDH was used as a loading control. In panels C, D and E, cells were treated with 1 µg/ml of LPS in the presence or absence of SP600125 (10 µM), a JNK inhibitor, for 48 hours. C: western blot analysis of IDO expression, D: mean±SEM C57BL/6 NOD IDO  !-actin LPS (µg/ml)          0         1            0        1 LPS (µg/ml)          0                1         C57BL/6 NOD IDO/!-actin ratio 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0  LPS       "        +       +        "        +        +                SP600125       "        "       +        "        "        +                C57BL/6 NOD IDO  !-actin IDO/!-actin ratio 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 LPS       "           +            +                       SP600125       "           "            +          LPS          "            +             +                       SP600125          "            "             +          Kynurenine level  (µg/ml) 7 6 5 4 3 2 1 0 A B C D E ! )(!ratio of densities of IDO to !-actin. E: the kynurenine levels indicate that LPS induced the expression of active IDO enzyme in NOD and C57BL/6 fibroblasts, which was mediated through JNK pathway. *denotes significant difference between cells from the same strain treated with LPS with or without SP600125 (p<0.05, n=3). **represents a significant difference between LPS-treated and non-treated cells from the same strain (p<0.001, n=3; refers to the number of independent experiments). ND: not detected.  3.3 DISCUSSION        Knowing that IFN#-induced IDO expression in many different cells including fibroblasts is important in controlling the number and viability of CD4+ and CD8+ T cells [232], here we asked the question of whether IFN# induces the expression of IDO in fibroblasts from an autoimmune diabetic mouse model.  We compared dermal fibroblasts from NOD and C57BL/6 mice in terms of: i) IFN#-induced IDO expression and enzymatic activity; ii) other IFN#-mediated pathways such as MHC I expression; iii) modulation of type I collagen expression; iv) IDO mRNA translational capacity; v) phosphorylation of STAT1, a key signal transducer in IFN#-mediated-IDO expression pathway; and finally, vi) ability of NOD cells to express IDO in response to an IFN#-independent stimulus such as LPS.   Our results showed that IFN# fails to induce IDO expression in dermal fibroblasts of NOD mice. As mentioned before, the NOD strain of mouse is a renowned animal model of autoimmune diabetes. This strain develops a spontaneous T cell-mediated destruction of pancreatic beta cells with a higher incidence in females compared to males. The onset of diabetes is about at 12 to 14 weeks of age in females and shortly after in male NOD mice [233]. In this study, we evaluated the effect of IFN# in expression and enzymatic activity of IDO in a selection ! ))!of prediabetic, diabetic, and non-diabetic NOD mice. The rationale for studying this group of mice was the previous observation that only DCs from prediabetic female NOD mice fail to express IDO following IFN# treatment. Our findings revealed that in contrast to what was reported for DCs, dermal fibroblasts from none of these age groups of NOD mice expressed IDO, and as a result no enzyme activity was found as evaluated by measurement of kynurenine level.  Considering the fact that in type 1 diabetes there is massive infiltration of T cells in pancreatic islets and expression of IDO has a T cell suppressive effect [232], it would be important to understand how IDO expression is involved in controlling islet inflammation.  Structurally, surrounding pancreatic islets, there exists a natural scaffold produced by fibroblasts [82]. Fibroblasts play a major role in improving the survival and functionality of islets. Here we hypothesized that failure of these fibroblasts to express IDO might also play a role in failure to suppress autoreactive T cells invading pancreatic islets. It is likely that IFN# released from activated T cells induces IDO expression in fibroblasts leading to a balance between tolerance and autoimmunity. Therefore, defective tryptophan catabolism in NOD fibroblasts along with those of DCs may be involved in the pathogenesis of autoimmune diabetes in the NOD mouse.   The MHC class I molecule is an abundant protein found at the cell surface of all nucleated cells. In fact, this molecule plays an important role in initiation of T cell mediated events [234]. The current study showed that IFN# is capable of inducing MHC class I expression in NOD dermal fibroblasts. Moreover, we found that IFN# decreased type I collagen expression in NOD dermal fibroblasts compared to that of control fibroblasts. The ability of IFN# to inhibit type I collagen synthesis has been previously reported [188]. It has been also been recently reported that this effect may be mediated through C/EBP!, a mediator of immune and inflammatory ! )*!responses, and independent of the JAK/STAT1 pathway (Figure 3.9). Indeed, IFN#-induced-phosphorylation of the extracellular signal-regulated kinase (ERK)1/2 mitogen-activated protein kinases  provokes phosphorylation and nuclear translocation of cellular C/EBP!. By binding to procollagen gene promoter, C/EBP! inhibits type I collagen expression [227]. We found that IFN# decreased collagen expression in both NOD and control fibroblasts. Our findings suggest that impaired responsiveness of NOD fibroblasts to IFN# is not due to lack or dysfunction of cytokine receptor.   To further explore the mechanism underlying defective tryptophan catabolism in NOD dermal fibroblasts, we asked whether the protein machinery system of the cells was malfunctioning. Fibroblasts were transduced with IDO gene using adenoviral vector.  In contrary to our observation after IFN# treatment, NOD fibroblasts were able to express IDO post transduction. Our results indicate that these cells were able to produce to functional IDO protein; we concluded that the defect in IFN#-mediated-tryptophan catabolism in NOD dermal fibroblasts was neither related to interaction of IFN# with its receptor nor gene transcription / translation mechanisms.   To get more insights into defective IDO expression in NOD fibroblasts, STAT1 phosphorylation was evaluated following IFN# treatment. We provided evidence that protein STAT1 did not get phosphorylated on tyrosine 701, impeding IFN#-induced- JAK/STAT1 pathway in NOD fibroblasts. These data were consistent with a previous observation of tolerogenic DCs in prediabetic NOD mice [214].  Thus, blockade of the JAK/STAT1 pathway in dermal fibroblasts from NOD mice is likely involved in the defect in IFN# induced-IDO expression.  ! *+!   Figure 3.9 IFN#-mediated MHC I expression and type I collagen repression.  IFN# induces MHC I expression via both JAK/STAT1 and canonical-NF%B pathways. IFN# downregulates type I collagen expression via the activity of C/EBP! protein, independent of conventional JAK/STAT1 pathway. C/EBP!: CCAAT/enhancer-binding protein !, COL-I: Type I collagen, ERK: Extracellular signal-regulated kinase, GAS: Gamma activating sequence, IKK: I%B kinase, IRF: Interferon regulatory factor, ISRE: Interferon-stimulated response element, PKR, Protein kinase R.  ! *"! Finally, we asked whether NOD dermal fibroblasts were able to express IDO through an IFN#-independent pathway. For this reason, we treated both C57BL/6 and NOD dermal fibroblasts with LPS. LPS has been shown to induce IDO expression in primary rat glial cultures without an increase in detectible IFN# [235, 236]. Although it is widely accepted that IFN# is an essential factor for IDO induction through the JAK/STAT1 pathway [230, 237], recent studies have revealed that IDO expression can be regulated by other inflammatory stimuli, including TNF-$ and LPS [238]. In this study, LPS was found to induce IDO expression through the activity of mitogen activated-protein kinase JNK (Figure 3.10); however, there is evidence that it could indirectly induce IDO expression e.g. via induction of type I IFNs through the canonical NF-%B pathway [160]. These data indicate that both IFN#'dependent and 'independent pathways can mediate IDO expression.  Our results showed that while IFN# fails to induce IDO expression in NOD dermal fibroblasts, these cells are able to express significant levels of active IDO enzyme in response to LPS treatment.  Confirming previous findings, our results also support the idea that LPS-induced IDO expression occurs via the JNK pathway.   ! *#!  Figure 3.10 LPS induces IDO expression via the activity of JNK and independent of JAK/STAT1 signaling pathway. Following TLR4 activation by LPS, a complex signaling cascade involving MyD88, IRAK, ERK, p38, and JNK kinase activation occurs. JNK activation directly activates transcription and nuclear translocation of AP-1 that is potential regulatory element of IDO. AP-1: Activator protein-1, ERK: Extracellular signal-regulated kinase, IKK: I%B kinase, IRAK: IL-1 receptor associated kinase, ISRE: Interferon-stimulated response element, MyD88: Myeloid differentiation primary response 88, TRAF6: TNF receptor associated factor, TRIF: TIR-domain-containing adapter-inducing interferon-!.  !"#$!"#$%&'()% (*+%,-"./%!"#$%AP-1 AP-1 AP-1 TLR4 MyD88 TRIF IRAK TRAF6 IKK IkB RelA JNK AP-1 0"'%P50 RelA P50 ! *$! In conclusion, the findings of this study provide evidence that IFN# fails to induce IDO expression and tryptophan catabolism in dermal fibroblasts from either female or male NOD mice regardless of the stages of diabetes. Moreover, our data suggest that this defect is neither due to IFN# receptor dysfunction nor protein expression machinery of these cells. Our findings further suggest that impaired STAT1 phosphorylation might be partially responsible for this failure. Despite the defect in JAK/STAT1 pathway, we showed that other IFN#-mediated pathways like IFN#-induced MHC I expression or IFN#-mediated inhibition of type I collagen production were still functional. As stated before, both of these pathways can be activated independent from JAK/STAT1. Therefore, the observed defect in this pathway would not interfere with other aspects of IFN#-mediated gene expressions and functions.  Although IFN# fails to induce tryptophan catabolism in NOD dermal fibroblasts, activation of an IFN#-independent pathway (such as LPS-mediated JNK pathway) can result in expression of active IDO enzyme.            ! *%!          Chapter 4:  EMBEDDING ISLETS IN A LIQUID  SCAFFOLD INCREASES  ISLET VIABILITY AND FUNCTION           ! *&!4.1 INTRODUCTION Replacement of the insulin producing cells by islet cell transplantation in selected individuals has been recognized as a promising treatment option for type 1 diabetes. This strategy leads to restoration of endogenous insulin production under the control of humoral and neuronal systems [7, 35, 49]. Successful islet transplantation trials began in early 1990s [68], raising hope that islet transplants would become the definitive treatment for restoring efficient insulin regulation. Unfortunately, international trials conducted around the world have demonstrated a constant decline of islet graft function post transplantation and therefore a low rate of successful long-term engraftment [70]. As mentioned before, currently human islet transplantation is limited by the shortage of islet donors and poor survival of the islets following transplantation [7, 35]. Even with the advent of the Edmonton protocol for islet transplantation introduced in 2000, islet graft function decreases over time so that by 5 years post-transplant less than 10% of the recipients remain insulin independent [7].  It is believed that both immune and non-immune factors are involved in reduction of beta cell mass and graft failure [71]. In addition, islet cell replacement using allogeneic grafts has shown short-term success, yet the recipients require systemic immunosuppressive medications, proven to be toxic and likely contributing to graft loss [7, 35]. It is estimated that a marked percentage of islets undergo apoptosis during the processes involved in islet isolation and transplantation [7, 104]. Therefore, effective strategies to reduce islet cell death during culture, transplantation and engraftment have the potential to improve the outcome. Re-establishment of islet interactions by restoring the ECM, lost during isolation procedures, is an attractive approach to improve the survival and function of the islets post transplantation. Our research group has previously developed a fibroblast populated collagen ! *'!matrix (FPCM), which significantly improves islet cell viability/function and reduces the marginal islet mass needed to correct hyperglycemia in mice with diabetes. Fibroblasts were incorporated to provide a favorable support for islets by producing various growth and angiogenic factors, while maintaining the integrity of the collagen matrix [217]. Although this composite was able to promote the islet graft survival and function, it was still prone to shrinkage, contraction and gradual biodegradation.  For this reason, we have developed and applied a novel bioengineered cross-linked-interpenetrating network of glycosaminoglycan and type I collagen (cross-linked collagen matrix, CCM). When populated with fibroblasts (FP-CCM), it would provide an optimal matrix biomimetic with reduced contraction and enhanced mechanical strength for the transplanted islets. Moreover, to avoid use of systemic immunosuppressants, we propose the use of IDO for generation of a local immunocompetent environment. We have previously shown that in an IDO-mediated microenvironment, T cells are not able to survive, proliferate or destroy the engrafted islets [209]. We have also previously demonstrated that by embedding islets in a collagen matrix populated with IDO-expressing fibroblasts (in which IDO expression was induced using an adenoviral vector), we could prevent the graft from alloimmune rejection in STZ-induced diabetic mice for more than 40 days [210]. The reason for seeing allogenic islet rejection around 40 days post transplantation was that IDO expression by fibroblasts gradually decreased in fibroblast due to the nature of adenoviral vector used.  Herein, we applied a more stable transduction method for induction of IDO expression in fibroblasts i.e. using a lentiviral vector. In this part of our study we also evaluated the viability and function of the islets embedded within the CCM, co-cultured with genetically modified fibroblasts in which IDO expression was transduced using a lentiviral vector bearing human IDO gene. ! *(!4.2 RESULTS 4.2.1 Cross-linked collagen matrix preserves normal morphology and insulin/glucagon expression in grafted islets. Isografts were prepared by embedding syngeneic Balb/c islets and normal fibroblasts within either collagen matrix (CM) or CCM. Figure 4.1A and 4.1B show islets embedded within the FP-CCM. The composite grafts were then transplanted under the kidney capsule of STZ-induced Balb/c diabetic mice (Figure 4.1C). Animals were monitored daily up to 100 days post transplantation.  At this time, the graft-bearing kidneys were retrieved and subjected to both H&E staining and immunostaining for insulin and glucagon expression. Histologic evaluation revealed a significant shrinkage of the islets when embedded within FPCM due to poor matrix biomimetic due to the contraction of the graft (Figure 4.1D). However, when islets were incorporated into the FP-CCM, healthy islets with normal spherical morphology due to lack of any contraction were observed (Figure 4.1E). Moreover, fibroblast cellularity was less within cross-linked matrix (Figure 4.1D and 4.1E). Immunostaining of islets in both FPCM and FP-CCM showed strong signals for insulin and glucagon 100 days following transplantation. However, the number of ! cells appeared to be lower in FPCM grafts and the $/! cell ratio was higher in these islets. suggesting more ! cell loss in FPCM grafts compared to FP-CCM grafts. Transplant recipients remained euglycemic during the course of the study.  ! *)! Figure 4.1 Histological assessment of engrafted islets embedded within fibroblast populated collagen matrix (FPCM) or fibroblast populated cross-linked collagen matrix (FP-CCM). Isografts were prepared by embedding syngeneic Balb/c islets and normal fibroblasts within either collagen matrix (CM) or cross-linked collagen matrix (CCM). A and B: Photomicrographs of islets embedded within the FP-CCM. C: The composite grafts were transplanted into renal subcapsular space of STZ-induced Balb/c mice with diabetes. D and E: Representatives of islet-FPCM and islet-FP-CCM isografts after 100 days posttransplantation, respectively. Paraffin-50 µm 50 µm 50 µm ! **!embedded sections of retrieved islet-FPCM and islet-FP-CCM composites grafts 100 days post-transplantation were subjected to H&E, and double immunofluorescence staining insulin (red) and glucagon (green). Lower panels show high magnification field of the area marked in the upper panels. FP-CCM promotes insulin and glucagon production while maintaining the $/! cell ratio. Size bar = 50 µm  4.2.2 Cross-linked collagen matrix improves the islet survival and viability To examine the viability of the islets within the FP-CCM, murine islets were incorporated within either FPCM or FP-CCM and cultured for up to 30 days. A group of islets was also co-cultured with normal C57BL/6 fibroblasts in regular two-dimensional (2D) cell culture condition as a control. Islet cell viability was then tested using a Live/Dead cell viability assay on days 1 and 30 post co-culture (Figure 4.2A). The result showed that the majority of islet cells died when cultured in 2D cell culture condition for 30 days (lower row). Although a smaller number of dead cells were found in islets embedded within FPCM on day 30 post co-culture (middle row), the best survival was achieved when islets where embedded within FP-CCM (upper row). Consistent with the panel A, Figure 4.2B represents the quantification of islet cell death at the above tested conditions.  Next, we evaluated islet cell survival and viability in the presence of IDO-producing fibroblasts, in which IDO expression was transduced using a lentiviral vector carrying a human IDO cDNA (Lenti-IDO). Briefly, murine islets were embedded within CCM either alone or co-cultured with normal or Lenti-IDO-transduced C57BL/6 fibroblasts for up to 30 days. A group of islets were cultured in 2D cell culture condition as a control group. A potential concern was the toxicity of lentiviral vector for the embedded islets; to address that we also included another ! "++!group in which islets were embedded within the CCM co-cultured with mock lentiviral vector (Lenti-Vect) transduced fibroblasts. Islet morphology was assessed during the course of experiment and islets were subjected to viability assay following 1, 15 or 30 days culture (Figure 4.3A). Figure 4.3B represents the quantification of islet cell death at the conditions shown in Figure 4.3A. The data show that embedding islets within CCM, both in the presence or absence of Lenti-IDO-transduced fibroblasts, appears to have no detectable negative effect on islet cell function and viability. As shown in Figures 4.3A and B, the number of dead cells is markedly higher in 2D cultured islets on days 1, 15 and 30 post-culture compared to those of islets embedded within CCM in the presence or absence of IDO producing fibroblasts. These data showed no significant difference between viability of the islets when cultured alone or co-cultured with Lenti-IDO fibroblasts within CCM. Moreover, the lentiviral vector itself appears not to be toxic to the islets because the Lenti-Vect transduced fibroblasts did not remarkably affect the survival of the islet cells when compared to normal fibroblasts. Finally, the conditioned media of islets cultured in standard conditons, islets alone in CCM and islets co-cultured with Lenti-IDO cells were collected at different time points for up to 30 days and subjected to kynurenine assay (Figure 4.3C). The data showed that kynurenine levels were markedly higher in Lenti-IDO fibroblasts populated islet composite grafts compared to islets embedded alone in the CCM matrix or in a regular culture dish, indicating IDO activity in fibroblasts incorporated within the CCM along with islets.   ! "+"!    Figure 4.2 Comparison of islet cell viability and survival within collagen and cross-linked collagen scaffolds. A: Mouse islets were embedded within fibroblast populated collagen matrix (FPCM)(middle row) or fibroblast-populated cross-linked collagen matrix (FP-CCM)(upper row) for up to 30 days. A group of islets was cultured in regular two dimensional (2D) culture dish (lower row). Islet cell viability was investigated using Live/Dead Viability/Cytotoxicity assay kit. B: Quantification of cell death in islets cultured in 2D dish, FPCM or FP-CCM for up to 30 days, reported as the average of red fluorescence intensity normalized for the islet area. * corresponds to significant difference from islets cultured in 2D at the indicated time point P<0.001, n=3; refers to the number of independent experiments).  0 5 10 15 20 !"#"$"D a y 1 Day 30 B/F  Live   Dead  Merged  A Day 1 Day 30 Day 1 Day30 Day 1 Day 30  2D         FPCM       FP-CCM  B Red fluorescence intensity/ Islet area !"!"! "+#!  Figure 4.3 Islet cell viability and survival within cross-linked collagen scaffolds at different conditions. A: Mouse islets were either embedded within CCM in the absence (ICCM) or along with Lenti-IDO-transduced fibroblasts (I-ICCM), Lenti-Vect-transduced fibroblasts (V-ICCM), normal C57BL/6 fibroblasts (B6-ICCM) or cultured in 2D culture dish (2D) or for up to 30 days. Islet cell viability was investigated using Live/Dead Viability/Cytotoxicity assay kit on days 1, 18 16 14 12 10 8 6 4 2 0 2D ICCM I-ICCM B6-ICCM V-ICCM Day 15 Day 30 B/F  Live   Dead  Merged  B/F  Live   Dead  Merged  Day 1 A 0 2 4 6 8 10 12 14 16 18 !" #" $" %" &"Day 1 Day 15 Day 30 2D       ICCM        I-ICCM      B6-ICCM     V-ICCM   Red fluorescence  intensity/ Islet area B !" !" !" !"# # # # C 0 2 4 6 8 10 12 14 16 18 1 5 10 15 20 25 30 Day  IICLC ICLC 2D Kynurenine (µg/ml)    B/F  Live   Dead  Merged  #"#" #"#"I-ICCM ICCM 2D ! "+$!15 and 30 post-culture. The first column on left shows morphology of a representative islet in each group under inverted bright field (B/F) microscope. Columns 2-4 show live cells (green), dead cells (red) and merged pictures under ultraviolet light, respectively. Optimal viability of the islets was achieved when they were embedded within FP-CCM matrix. Co-culture of the IDO producing fibroblasts transduced with lentiviral vector was nontoxic to the embedded islets. B: Quantification of cell death in islets cultured in the conditions of panel C, reported as the average of red fluorescence intensity normalized for the islet area. *, # and ¶ indicate significant difference from islets cultured in 2D condition on day 1, 15 and 30 postculture, respectively (p< 0.001, n=3; refers to the number of independent experiments). Data are reported as mean±SD. C: Conditioned media of islets cultured in 2D, CCM (ICCM) and Lenti-ID-populated CCM (I-ICCM) were subjected to kynurenine assay at different time points to indicate the IDO activity. Tryptophan was markedly catabolized in the presence of IDO-producing fibroblasts vs. other groups (p<0.001, n=3; refers to the number of independent experiments). Data are reported as mean±SD.  4.2.3 Cross-linked collagen matrix preserves islet insulin secretory function and reduces fibroblast proliferation rate To investigate the effect of FPCM and FP-CCM on islet function, islets after being embedded within the three dimensional composites were retrieved on days 1, 15, or 30 post co-culture and subjected to glucose-stimulated insulin secretion assay. The insulin secretory capacity of islets was assessed by comparing the percentage of cellular insulin released in low (1.67 mM) versus high (16.7 mM) glucose media. We observed a significant decrease in glucose-stimulated insulin secretion in islets cultured in the regular 2D setting for 15 and 30 days. On the other hand, glucose responsiveness and insulin secretory capacity were preserved to levels similar to that at ! "+%!day 1 of culture when islets were embedded within either CM or CCM for 15 days (Table 4.1; Figure 4.4A). After 30 days, however, islets embedded within FPCM started to show a mild but significant decrease in glucose responsiveness compared to day 1 (stimulation index: 2.02±0.18 vs. 2.92±0.21), whereas FP-CCM preserved glucose stimulated insulin secretion in islets even after 30 days culture (stimulation index: 2.80±0.38 vs. 3.22±0.23) (Table 4.1; Figure 4.4A). A potential concern regarding incorporation of fibroblasts within composite grafts is over-proliferation of these cells, which may jeopardize islet function and survival. To gain a perspective on this issue, we assessed fibroblast proliferation in 3 different conditions:.2D culture in tissue plates, CM, or CCM. The results (Figure 4.4B) showed that embedding fibroblasts within CCM significantly reduced, though did not eliminate, fibroblast proliferation compared to the two other conditions (2.3±0.31 vs. 4.5±0.41 vs. 8.8±0.80-fold after 30 days culture in CCM, CM or 2D, respectively)(Figure 4.4B).            ! "+&!        Table 4.1 Glucose-stimulated insulin secretion assay. Mouse islets were embedded within fibroblast-populated collagen matrix (CM) or fibroblast-populated cross-linked collagen matrix (CCM) or cultured in two dimensional culture dish (2D) for up to 30 days. Islets were retrieved on days 1, 15, or 30 post co-culture and glucose-stimulated insulin secretion assessed. Data are presented as mean±SD.        Basal Stimulated Stimulation IndexCCM !"#$%#"&! '"#$%#"(! !"))%#")!CM !"!*%#"+! '"*+%)")* )"')%#")&2D )"'&%#"+, *"*&%)")+ !"##%#")*Basal Stimulated Stimulation IndexCCM !"&!%)"$! '",'%)"$! )"''%#"!+CM )"(&%#"$& ,"*,%#"$# )"+,%#"!!2D &"*)%#")! )"!!%#")( &")'%#")&Basal Stimulated Stimulation IndexCCM )"&(%#"+& ,"&)%)"#! )"*%#"!*CM !",(%#"$( ("!*%#",+ )"#)%#"&*2D #"'(%#"!$ #"*'%#"$& #"*'%#"&'Day 30Day 1Day 15! "+'!   Figure 4.4 Islet insulin secretory function and fibroblast proliferation within collagen composite scaffolds.  A: Glucose-stimulated insulin secretion from islets following over 30 days of culture in fibroblast populated-CM or -CCM compared to standard culture condition (2D). B: Fibroblast proliferation rate was evaluated within 2D culture and CM and CCM matrices. * corresponds to significant difference between indicated groups (p<0.001, n=3; refers to the number of independent experiments). Data are presented as mean±SD.     ! "+(!4.2.4 Length of IDO expression and effect on tryptophan catabolism in Lenti-IDO-transduced fibroblasts Another concern regarding composite grafts with Lenti-IDO-transduced fibroblasts was the stability of IDO expression in transduced fibroblasts when embedded within the matrix. To answer this, we incorporated different numbers of Lenti-IDO-transduced fibroblasts (50)103,100 )103,150)103, 200)103 cells) within either CM or CCM. Lenti-Vect transduced fibroblasts populated CM and CCM were used as control groups. The levels of L-kynurenine were measured at different time points for up to 75 days, for evaluation of IDO activity (Figure 4.5). The data show that the kynurenine levels were significantly higher in Lenti-IDO populated CM (Figure 4.5A) and CCM (Figure 4.4C), compared to that of Lenti-Vect transduced fibroblast populated CM (Figure 4.5B) and CCM (Figure 4.5D), respectively (p<0.01, n=3) at each time point for up to 75 days. Furthermore, IDO mRNA expression was evaluated in composite matrices composed of 200)103 fibroblasts using RT-PCR analysis for up to 90 days (Figure 4.5E). Consistent with the enzymatic activity of IDO, RT-PCR assay showed that IDO mRNA expression was significant and persistent in Lenti-IDO-transduced fibroblasts embedded within both CM and CCM matrices (Figure 4.5E, upper row). In addition, no IDO mRNA expression was detected in FPCM or FP-CCM composed of Lenti-Vect transduced fibroblasts (Figure 4.5E, lower row). Collectively, these data confirmed that both CM and CCM were harmless to the fibroblasts and the lentiviral transduction method resulted in stable tryptophan catabolism and IDO mRNA expression for at least 90 days. !! "+)! Figure 4.5 Lasting effect of tryptophan catabolism in Lenti-IDO-transduced fibroblasts embedded within collagen scaffolds. L-kynurenine levels were measured in conditioned media of the Lenti-IDO fibroblast populated CM (IFCM) (A) and CCM (IFCCM) (C) for up to 75 days. Lenti-Vect-transduced fibroblasts populated-CM (VFCM) (B) and-CCM (VFCCM) (D) were used as control groups respectively. Lenti-IDO-transduced fibroblasts (200)10^3) were 0 5 10 15 20 25 30 3 9 15 24 31 42 47 53 60 68 75 !"##$%&'(!"##$%)''(!"##$%)&'(!"##$%*''(!"#$%&#'#&(µ)*+,((((Day 0 5 10 15 20 25 30 3 9 15 24 31 42 50 57 66 75 !"#$%&'(!"#$%)''(!"#$%)&'(!"#$%*''(!"#$%&#'#&(µ)*+,((((Day IFCM                    IFCCM VFCM               VFCCM IDO GAPDH   IDO GAPDH 15  30  45  60  75  90      15  30  45  60  75  90 0 5 10 15 20 25 3 9 15 24 31 42 50 57 66 75 !"#$%&'(!"#$%)''(!"#$%)&'(!"#$%*''(!"#$%&#'#&(µ)*+,((((Day !"#"$!"$#"%!"%#"&" '" $#" %(" &$" (%" ()" #&" *!" *+")#",-../0#!",-../0$!!",-../0$#!",-../0%!!"B A D C !"#$%&#'#&(µ)*+,((((Day E ! "+*!embedded within CM (IFCM) or CCM (IFCCM), and IDO mRNA expression was evaluated in these cells for up to 90 days (E, upper row). Lenti-Vect-transduced fibroblasts populated-CM (VFCM) and -CCM (VFCCM) (E, lower row) were used as control groups, respectively.   4.3 DISCUSSION The Edmonton protocol, introduced in 2000, achieved approximately 80% successful insulin independence for the first year after receiving final islet infusion [76]. This protocol suggests a steroid-free immunosuppression along with a very high number of islets provided by multiple infusions [7]. Unfortunately, longer-term results were less encouraging. Islet function decreases over time so that by 5 years post-transplantation, less than 10% of the recipients remain insulin-independent [71].  Myriad overlapping forces contribute to decrease the islet graft function and survival after transplantation. During islet isolation, which takes a number of hours, islets are exposed to various mechanical, enzymatic, osmotic and ischemic stresses. Then they are transplanted to a foreign environment where they are exposed to hypoxia, immune attack, systemic toxins and immunosuppressive drugs [104, 239]. Indeed, islets must adapt to an environment lacking vascularization, innervation and the native peri-insular ECM [94]. It has been shown that interactions with ECM play an important role in preservation of spherical morphology, survival and insulin secretory function of mature, intact islets while transmitting a variety of biochemical and biomechanical signals mediating key aspects of islet cell physiology. Interestingly, incompletely isolated islets that have remnants of ECM have been shown to have significantly ! ""+!decreased apoptosis and improved insulin response compared to purified and completely isolated islets [200]. ECM is also shown to act as a reservoir for growth factors and cytokines and protects them from degradation while presenting them to their receptors on cells. Loss of peri-insular ECM and peri-vascular basement membrane during the islet isolation process can significantly disrupt ECM-mediated interactions and result in apoptosis [94, 240]. Various types of matrices and scaffolds have been proposed to promote islet cell viability and function including but not limited to synthetic polymers [152, 153], various types of collagen [148, 152, 241], Matrigel [154], and small intestinal submucosa [155, 242]. Like any biodegradable scaffolds, these matrices undergo gradual degradation. Moreover, it takes a long time to demonstrate the safety of new synthetic polymers and translate them to clinical application [151]. Other strategies to protect engrafted islets include application of angiogenic, anti-apoptotic, and growth factors [243-246]. However, these approaches usually require direct gene transfer to the islets that can negatively affect islet viability and survival [217]. Our research group has previously introduced a novel composite scaffold for islets consisted of syngeneic fibroblasts embedded within collagen matrix that attempts to address both of these issues (i.e. ECM and growth factors) [217]. Collagen scaffold provides a favorable ECM for islets whereas fibroblasts support and maintain the matrix integrity and furthermore provide islet growth factors. Nevertheless, this matrix is still not an optimal scaffold for the islets: collagen scaffold has a tendency to contract when populated with fibroblasts due to both mechanical forces and forces caused by the cells. Furthermore, collagenase produced by cells like fibroblasts is able to remodel and degrade collagen scaffolds. Disintegrated collagen particles have been shown to induce proliferative, fibrotic and immune responses [247]. Like previously suggested scaffolds, the FPCM is a ! """!preformed solid scaffold and therefore, the engraftment of the islets should be prepared several hours before transplantation that could compromise islet cell viability due to hypoxic and mechanical stresses. Moreover, when transplanted underneath the renal capsule, the composite graft does not immediately integrate with the surrounding tissue. In this study, we have applied a novel cross-linked interpenetrating network of glycosaminoglycan and type I collagen, which has previously demonstrated the potential to diminish cell proliferation and contraction without affecting cell viability [151]. Additionally, we have shown previously that glutaraldehyde cross-links and polyvinylalcohol-borate networks within this bioengineered matrix prevent enzymatic attack to digest the collagen fibrils [151]. The proposed matrix is a liquid matrix that solidifies at 37°C and could integrate with the surrounding tissue on application. Additionally, cells could be embedded before casting a few minutes before transplantation. The findings of the present study show clearly that our novel bioengineered scaffold is nontoxic to the embedded islets. Histological assessment of the engrafted islets confirmed that even after 100 days following syngeneic transplantation, islets maintain their spherical morphology within the FP-CCM with minimal detectable contraction or change in $/! cell ratio. Furthermore, our results revealed significantly enhanced islet cell survival, insulin responsiveness and reduced fibroblast proliferation rate after being embedded within FP-CCM for 30 days as compared to regular 2D and CM culture environment. Our research group has previously shown that local IDO expression delivered by fibroblasts transduced with adenoviral vector carrying the IDO gene, prevents cellular and humoral alloimmune responses against islets and markedly prolongs islet allograft survival (for an average of about 42 days) without systemic immunosuppressive treatments [210]. We applied a more stable gene transfer method for induction of IDO expression in fibroblasts embedded ! ""#!within the CCM. The viability and survival of the islets were then evaluated in the presence of Lenti-IDO-transduced fibroblasts when both embedded within the CCM matrix. The results show that after 30 days in culture, not only does embedding islets within CCM matrix remarkably improves islet cell viability, but also transduction of fibroblasts with Lenti-IDO does not compromise their survival; the viability assessment of islets in the presence of Lenti-Vect transduced fibroblasts confirmed that the lentiviral transduction itself is harmless to the islet cells viability and insulin/glucagon production. In addition, we showed that IDO expression in fibroblasts embedded within the matrix is stable for at least 90 days. This finding opens new hopes for more prolonged graft protection against rejection.  Taken together, this part of our study clearly shows that our novel CCM, when populated with fibroblasts, not only has the beneficial properties of the collagen scaffolds but also addresses two major concerns with FPCM that are poor biomimetic and gradual disintegration. Our result shows that the FP-CCM provides an improved mechanical strength maintaining natural structure, viability and functionality of the embedded islets. Moreover, incorporation of IDO-producing fibroblasts within the CCM does not harm the viability or function of the islets. This promising finding offers a valuable new approach to improve the outcome of islet transplantation. Further investigations are required to evaluate the functionality of CCM populated with IDO-producing fibroblasts in allogeneic transplantation models.      ! ""$!        Chapter 5:  IMMUNOPROTECTION AND FUNCTIONAL IMPROVEMENT OF ALLOGENEIC ISLETS IN DIABETIC MICE, USING A STABLE IDO-PRODUCING SCAFFOLD      ! ""%!5.1 INTRODUCTION Interest has piqued in islet transplantation as a treatment option for type 1 diabetes. Progress achieved through improved islet isolation techniques, avoidance of glucocorticoids for immunosuppression and transplantation of adequate islet mass have led to improved outcomes in diabetic recipients with hypoglycemic unawareness [7, 35, 49, 80]. As discussed before in detail, despite the many advances in islet transplantation, several significant issues still limit this promising procedure and its outcomes. Necessity of multiple donors, sub-optimal islet yield and quality using current islet isolation protocols, side effects associated with immunosuppressive drugs to prevent allorejection, poor engraftment and immediate post-transplant islet loss, and difficulties in maintaining long-term insulin independence are amongst the main challenges in clinical islet transplantation [64, 71, 239, 248, 249].         Pancreatic islets compose a small portion of the whole pancreas; the isolation process aims to remove the exocrine tissue through enzymatic digestion of the extracellular matrix (ECM) and mechanical separation followed by density gradient [71]. ECM plays an important role in islet survival [250] by providing stored vital growth factors and cytokines to the islets and mediating specific biochemical and mechanical signals that control cell migration, differentiation and survival [68, 251]. Islet isolation digests peripheral ECM and peri-insular basement membrane interrupts oxygen and nutrient delivery, and leads to anoikis-like apoptosis [240, 250, 252, 253], compromising islet cell survival and reducing islet yield for transplantation. Although isolation and transplantation do not completely upset the islet-ECM interaction, they lead to disruptions that may markedly affect islet engraftment and performance [94]. Therefore, replacement of the ECM is an attractive approach to improve the long-term survival of islets.  ! ""&!      We have previously developed a composite scaffold (i.e. the fibroblast populated collagen matrix; FPCM) that significantly improved islet cell viability and function and reduced the islet mass needed to achieve euglycemia in STZ-induced diabetic mice [217]. Collagen is a surrogate for the natural ECM of islets while fibroblasts provide a favorable support for them by producing various growth and angiogenic factors and maintaining the scaffold structure [217]. However, the FPCM was prone to gradual biodegradation and contraction, negatively affecting the long-term survival and function of the islets [241]. Recently, we developed and applied a novel bioengineered, cross-linked, interpenetrating network of glycosaminoglycan and type I collagen (cross-linked collagen matrix, CCM) [62, 241]. When fibroblasts were populated within CCM (FP-CCM), the scaffold provided enhanced mechanical strength and reduced contraction for the transplanted islets both in vitro and in vivo in a mouse model of syngeneic islet cell transplantation [62].           Despite their value in preventing islet allo-rejection, immunosuppressants directly contribute to !-cell toxicity and islet death [71]. Although the Edmonton protocol demonstrated the value of a corticosteroid-free regimen [80], combinations of other immunosuppressants such as tacrolimus and sirolimus may still decrease !-cell engraftment, survival, function and proliferation [113, 254]. One option to avoid using systemic immunosuppressive regimens is to locally induce an immunosuppressive factor, IDO, the first and rate-limiting enzyme in the tryptophan catabolism pathway [255]. This enzyme generates a microenvironment, low in tryptophan and rich in kynurenine metabolites, that is selectively toxic to T cells. We have previously shown that in such an environment, while allogeneic islets remain intact, T cells are not able to survive, proliferate or destroy the engrafted islets [209, 255]. Using the FPCM, we demonstrated that transient IDO transduction of the incorporated fibroblasts using an adenoviral ! ""'!vector prolongs the survival of allogeneic islets embedded within this scaffold up to six weeks [210].  The purpose of this study was, therefore, to overcome two main difficulties that we encountered in our previous research: (i) the gradual biodegradation of collagen and contractibility of FPCM that could compromise long-term transplant outcome [151]; and (ii) the transient IDO expression in fibroblasts, limiting the period of graft survival [170].  To overcome these problems, herein, we have applied: (i) a novel scaffold that offers an optimized biomimetic matrix with improved mechanical properties; and (ii) a recently constructed lentiviral vector bearing an IDO-expressing gene to generate more stable IDO expression in fibroblasts. Using these systems, we examined whether local stable IDO expression by FP-CCM prolongs islet allograft survival in a STZ-induced diabetic mouse model.  5.2 RESULTS 5.2.1 IDO-transduced fibroblasts embedded within CCM express IDO.  C57BL/6 fibroblasts transduced with a lentiviral vector carrying a human IDO cDNA (Lenti-IDO) were embedded within the CCM for up to 14 days. Lenti-IDO fibroblast-populated CCM was subjected to Live/Dead® Viability/Cytotoxicity assay on day 14 post culture. As shown in Figure 5.1A, the majority of fibroblasts were viable when embedded in CCM. Immunostaining of the composite scaffolds 14 days post-culture showed that embedded Lenti-IDO fibroblasts within CCM were able to express IDO while no IDO expression was present in Lenti-Vect fibroblasts. (Figure 5.1B).  ! ""(!    Figure 5.1 IDO expression in transduced fibroblasts embedded within cross-linked collagen matrix (CCM). A: Live/Dead® Viability/Cytotoxicity assay was performed to evaluate the viability of Lenti-IDO transduced fibroblasts embedded within the CCM following 14 days of culture. B: Immunofluorescence staining of IDO protein (red) counterstained with DAPI (blue) in composite Lenti-IDO- and Lenti-Vect- fibroblast populated (FP)-CCM following 14 days culture. IDO is highly expressed in Lenti-IDO fibroblasts while no IDO expression is detected in Lenti-Vect fibroblasts. Size bars = 50 µm; B/F: Bright field.     Lenti-Vect     Lenti-IDO B Merged Merged A Live Dead B/F IDO DAPI 50µm 50µm 50µm 50µm 50µm 50µm ! "")!5.2.2 Local IDO expression suppresses proliferation of diabetic splenocytes co-cultured with allogeneic islets. We next evaluated the suppressive effect of IDO on the proliferation of splenocytes isolated from STZ-induced diabetic C57BL/6 mice, when co-cultured with allogeneic Balb/c islets for 72 hours (Figure 5.2A). Splenocytes proliferated markedly in response to stimulation by allogeneic islets; however, local IDO expression delivered by lentiviral vector to co-cultured fibroblasts significantly reduced the proliferation of stimulated splenocytes. The suppressive effect of IDO was markedly reversed in the presence of 1-MT, a competitive IDO inhibitor (IDO+1-MT). Flow cytometric analysis (Figure 5.2B) showed about 67% change in the proliferation rate of splenocytes when co-cultured with allogeneic islets in the presence of IDO-producing fibroblasts (10.3±0.9%) compared to that of control (31.7±3.8%). That the suppression of proliferation was due to IDO activity was confirmed by almost complete reversal of the splenocyte proliferation rate in the presence of 1-MT (25.0±2.3%). ! ""*! Figure 5.2 IDO-expressing fibroblasts suppress the proliferation of alloantigen-stimulated mouse splenocytes. A: Photomicrographs of diabetic C57BL/6 splenocytes following 72 h co-culture with allogeneic (Balb/c) islets in the presence or absence of syngeneic fibroblasts transduced with lentivirus to express IDO. Size bars=50µm. Red arrows indicate co-cultured fibroblasts. B: Proliferation rates of CSFE-stained splenocytes stimulated with allogeneic islets for 72 h in the presence or absence of IDO-expressing fibroblasts and measured by flow cytometry. 1-MT was used as an inhibitor of IDO activity. Lenti-IDO transduced fibroblasts suppressed splenocyte proliferation and this effect was reversed by addition of 1-MT, confirming the suppressive effects of IDO expression. Red arrows indicate fibroblasts in culture. C: presentation of data as mean±SD. * corresponds to a statistically significant difference between indicated groups (P<0.001; n=3). 1-MT: 1-methyl tryptophan, SPL: splenocytes    Control                 IDO             IDO+ 1-MT      A B 0 5 10 15 20 25 30 35 40 !"#$%"&'CFSE !"#$%"&' ()*' '()*+',-./!* * Proliferating SPL (%) SSCA     Control                IDO             IDO+ 1-MT      50µm 50µm 50µm C ! "#+!5.2.3 Cross-linked collagen matrix improves viability and function of cultured murine islets. Murine islets were embedded within CCM either alone or co-cultured with non-transduced or Lenti-IDO transduced C57BL/6 fibroblasts for up to 30 days. A group of islets were cultured in a two dimensional (2D) culture dish as a control group. As another control for the potential toxicity of our lentiviral vector to embedded islets, we also included a group in which islets embedded within the CCM were co-cultured with fibroblasts transduced with a mock lentiviral vector (Lenti-Vect). Histological analysis revealed that embedding islets within CCM in the presence or absence of Lenti-IDO transduced fibroblasts did not affect insulin and glucagon expression (Figure 5.3A). Furthermore, CCM markedly improved islet viability compared to that of 2D culture alone as shown by higher percentage of cleaved caspase-3 expression in the latter, on days 15 and 30 post culture (Figure 5.3A and C). These data showed no significant difference in insulin or glucagon expression, and the viability of islets when cultured alone or co-cultured with Lenti-IDO fibroblasts within CCM was also not different. Moreover, the lentiviral vector itself appears to be harmless to the function and viability of islets since Lenti-Vect fibroblasts did not noticeably affect insulin, glucagon or caspase-3 expression in islet cells compared to those of normal fibroblasts. Embedding islets within the CCM was associated with significantly higher !/$ cell ratio than 2D cultured islets (Figure 5.3B) suggesting that the CCM likely increases ! cell viability in islets during culture.   ! "#"! Day 1 Day 15 Day 30 0 5 10 15 20 VCLC B6CLC IICLC ICLC 2D Day 30 Day 15 Day 1 0 5 10 !"#"$"%"&"2D ICCM I-ICCM B6-ICCM V-ICCM * * * # # # !/" cell ratio B A 2D ICCM I-ICCM B6-ICCM V-ICCM % Active caspase-3 +  insulin+ islet cells  C * # 2D  ICCM  I-ICCM  B6-ICCM  V-ICCM Ins Gluc DAPI Ins Casp DAPI Ins Gluc DAPI Ins Casp DAPI Ins Gluc DAPI Ins Casp DAPI * * * # # # * # ! "##!Figure 5.3 Lenti-IDO fibroblast populated cross-linked collagen scaffold preserves islet insulin and glucagon expression while reducing caspase-3 expression in ! cells in vitro. Islet grafts were harvested and fixed in paraformaldehyde on days 1, 15 and 30 post culture. A: Paraffin-embedded sections were subjected to double-immunofluorescence staining for insulin (red) / glucagon (green), and insulin (red) / cleaved caspase-3 (green) on days 1, 15 and 30 post culture. Size bar= 50µm. Note the apparent increase in the number of caspase 3-positive cells over time in two dimensional (2D) cultured islets (yellow arrow) compared to no caspase-3 expression in other groups. Cross-linked matrix, local IDO expression or lentiviral vector did not negatively affect islet insulin/ glucagon expression. B: The !/$ cell ratio was calculated by counting insulin- and glucagon-expressing cells in each islet, in a minimum of 20 islets per condition. Data are presented as mean±SD. Note that islets embedded within cross-linked collagen scaffold have higher !/$ cell ratio compared to 2D cultured islets. C: Active caspase-3 positive ! cells were counted in each islet (minimum of 20 islets per condition). Data are reported as the percentage of cleaved caspase-3+ insulin+ islet cells (mean±SD). Note that the number of caspase-3-expressing ! cells in significantly higher in 2D group compared to other conditions at each time point. * and # indicate significant difference versus 2D cultured islets  at each time point (P<0.01). 2D: regular two dimensional culture dish, ICCM: islets embedded alone in CCM, I-ICCM: islets embedded in Lenti-IDO fibroblast populated-CCM, B6-ICCM: islets embedded in normal C57BL/6 fibroblast populated CCM, V-ICCM: islets embedded in Lenti-Vect fibroblast populated-CCM, Ins: insulin, Gluc: glucagon, Casp: cleaved caspase-3.! "#$!5.2.4 Local IDO expression delivered by lentiviral vector improves islet transplantation outcome.  To investigate the local immunosuppressive effect of IDO, composite islet grafts were prepared by embedding Balb/c islets within CCM populated with C57BL/6 fibroblasts transduced with either Lenti-IDO, control Lenti-Vect or non-transduced control fibroblasts. Another control group received islets only. Islets were transplanted into STZ-diabetic, allogeneic (C57BL/6) recipients. The Kaplan-Meier survival curve derived from blood glucose measurement showed a significant prolongation of graft survival in IDO-expressing graft-recipients (51.0±2.9 days) compared to control grafts (Lenti-Vect populated CCM) that were rejected within 2 weeks after transplantation (12.0±1.1 days; Figure 5.4A). The mean survival duration of Lenti-Vect populated CCM was not significantly different from those of islet CCM grafts with untreated fibroblasts (12.6±1.0 days) or without fibroblasts (11.6±1.5 days), as well as islet-alone grafts with no matrix (11.0±1.0 days) (Figure 5.4B). Figures 5.4C and D show blood glucose levels in recipients of Lenti-Vect- and Lenti-IDO transduced grafts, respectively.  ! "#%!  Figure 5.4 Islet grafts survival after transplantation. Allogeneic islets were embedded within CCM populated with either IDO gene transduced (Lenti-IDO Fib), Mock vector transduced (Lenti-Vect Fib), or untreated C57BL/6 (UT Fib) fibroblasts and transplanted under the kidney capsule of STZ-induced diabetic mice. Some mice received islets alone embedded in CCM or untreated islets with no matrix. Blood glucose levels were measured in graft recipients at least twice a week. Diabetes was considered as two consecutive blood glucose levels " 20mM. A: Kaplan-Meier islet allograft survival curve. B: Blood glucose levels in recipients of Lenti-IDO transduced islet grafts and Lenti-Vect transduced islet grafts (control group).  Lenti-IDO Fib/ Islet/ CCM: islets embedded in IDO gene-transduced fibroblast populated-CCM. Lenti-Vect Fib/ Islet/ CCM: islets embedded in mock lentiviral vector-transduced fibroblast populated-CCM, UT Fib/ Islet/ CCM: islets embedded in untreated fibroblast populated-CCM, Islet/ CCM: islets embedded alone in CCM.   0 5 10 15 20 25 30 35 0 3 7 11 14 18 20 25 29 32 35 38 40 42 47 48 49 50 51 54 55 56 0 5 10 15 20 25 30 35 0 3 7 11 14 18 20 25 29 32 35 38 40 42 47 48 49 50 51 54 55 56 A Lenti-IDO graft Lenti-Vect graft C B Lenti-IDO Fib/ Islet/ CCM (n=5) Lenti-Vect Fib/ Islet/ CCM (n=3) UT Fib/ Islet/ CCM (n=3) Islet/ CCM (n=3) Islet (n=3) Islet allograft  survival (%) Days after transplant 100    50    0 0       20             40      60 Days after transplant Days after transplant BGL (mM) BGL (mM) ! "#&!5.2.5 Local expression of IDO preserves insulin expression and reduces immune cell infiltration.   To examine histopathological changes in the islet allografts, graft-bearing kidneys and draining lymph nodes were harvested at different time points. Islet grafts were stained with H&E. The islets in Lenti-Vect transduced (control) grafts were partially destroyed by 2 weeks post-transplant (Figure 5.5A and B) whereas islet architecture was well preserved in the IDO-expressing grafts for about 7 weeks (Figure 5.5C-F). Control grafts were highly infiltrated with mononuclear cells as early as week 2 (Figure 5.5A and B); however, Lenti-IDO transduced grafts remained intact with almost no detectable immune infiltration by 2 weeks (Figure 5.5C and D). Graft sections show that some immune cells began to surround islets in IDO-expressing grafts at 7 weeks after transplantation (Figure 5.5E and F), and this infiltration increased markedly by 8 weeks post transplant. (Figure 5.5G and H).  Graft-bearing kidneys were also subjected to double-immunofluorescence staining for insulin and CD3. As IDO is known to increase the number of Tregs, we also investigated the infiltration of FOXP3+ cells into the grafts and draining lymph nodes at different time points. Graft draining lymph nodes were stained for FOXP3.  Immunostaining for insulin and CD3 revealed very few insulin-producing cells but extensive infiltration of T cells in Lenti-Vect grafts (Figure 5.6A). The immunostained sections show that at ~2 weeks (Figure 5.6B) and 7 weeks post transplantation, Lenti-IDO grafts had strong staining for insulin, indicating preservation of functional beta cell mass, and confirmed the minimal immune cell infiltration at week 7 seen by H&E (Figure 5.6C). Massive T cell infiltration into the IDO-expressing grafts started around week 8 post transplantation (Figure 5.6D).  ! "#'!    ~Week 2 Lenti-Vect graft A B ~Week 2 ~Week 7 ~Week 8 Lenti-IDO graft C D E F G H ! "#(!Figure 5.5 Evaluation of islet structure and immune cell infiltration in composite islet grafts. Graft-bearing kidneys were retrieved at different time points and subjected to H&E staining. A, B: Lenti-Vect transduced composite islet graft at ~2 weeks post-transplant. C, D: Lenti-IDO transduced composite islet graft at ~2 weeks post-transplant. E, F: Lenti-IDO transduced composite islet graft at ~7 weeks post-transplant. G, H: Lenti-IDO transduced composite islet graft at ~8 weeks post-transplant. Note that cellular infiltration into the graft started in Lenti-Vect grafts as early as week 2 post-transplant, whereas delayed until week 7 post-transplant in Lenti-IDO grafts. Size bar = 20 µm.  Lenti-Vect graft: islets embedded in mock lentiviral vector-transduced fibroblast populated-CCM. Lenti-IDO graft: islets embedded in IDO gene-transduced fibroblast populated-CCM.     Although the FOXP3 immunostaining revealed very few FOXP3+ cells at the graft site and in draining lymph nodes in Lenti-Vect grafts at ~ week 2 post transplantation (Figure 5.6E and I respectively), there were markedly more infiltrating FOXP3+ cells in IDO-expressing grafts at the same time point (Figure 5.6F and J). At about 7 weeks post-transplant the number of FOXP3+ cells decreased significantly at both the graft site and draining lymph nodes (Figures 5.6G and K respectively). By the time of graft rejection, the number of FOXP3+ cells was comparable in Lenti-Vect (~ week 2) and Lenti-IDO (~ week 8) groups both at the graft site and in draining lymph nodes (Figure 5.6H and L respectively). These findings were confirmed by quantification of the number of graft infiltrating FOXP3+ cells (Figure 5.6M) suggesting that local IDO expression delivered by lentiviral vector prevents islet allograft rejection via prevention of T cell infiltration into the graft while increasing the number of FOXP3+ cells at the graft site.  ! "#)! FOXP3, DAPI ~Week 2 ~Week 7 ~Week 8 Lenti-IDO group Ins, CD3, DAPI FOXP3, DAPI ~Week 2 Lenti-Vect group Graft Lymph node H A E I B F J C G K D L !" #!" $!" %!" &!"Lenti-IDO 8Wk Lenti-IDO 7Wk Lenti-IDO 2Wk Lenti-Vect 2Wk * P=0.03 * P=0.006  Number of intra-graft FOXP3+ cells M ! "#*!Figure 5.6 Characterization of intra-graft and –draining lymph nodes infiltrating immune cells. Graft-bearing kidneys were retrieved at different time points and either double immunostained for CD3 and insulin or single immunostained for FOXP3. Draining lymph nodes were also retrieved and stained for FOXP3. A-D: Insulin expression and T cell infiltration in control (Lenti-Vect transduced) islet grafts at ~2 weeks and IDO-expressing (Lenti-IDO transduced) islet grafts at ~2, 7 and 8 weeks post-transplant respectively. E-H: Intra-graft FOXP3+ cells in control (~week 2) and IDO-expressing grafts (~week 2, 7 and 8) following transplantation respectively. I-L: FOXP3+ cells infiltrating draining lymph nodes of control graft recipients (~week 2) and IDO-expressing graft recipients (~week 2, 7 and 8) following transplantation. Note that T cell infiltration into the graft started in control group around week 2 post-transplant whereas delayed until week 7 in IDO-expressing grafts. M: Quantification of intra-graft FOXP3+ cells (n=3, refers to the number of independent experiments). Data are presented as mean±SD. Size bar = 20 µm.  Lenti-Vect group: Recipients of islets embedded in mock lentiviral vector-transduced fibroblast populated-CCM. Lenti-IDO group: Recipients of islets embedded in IDO gene-transduced fibroblast populated-CCM, Vect 2W: Lenti-Vect graft at week 2, IDO 2W: IDO-expressing graft at week 2, IDO 7W: IDO-expressing graft at week 7, IDO 8W: IDO-expressing graft at week 8.   5.2.6 Long-term expression of IDO in IDO-gene transduced islet grafts. To get insight to possible mechanisms underlying the decline in IDO-mediated protection against rejection at 8 weeks post-transplant, we measured IDO transgene expression in the Lenti-Vect grafts at 2 weeks post-transplant and in the Lenti-IDO grafts at 7 weeks post-transplant. Expression of IDO mRNA was easily detectable in Lenti-IDO expressing grafts while no IDO mRNA was detected in Lenti-Vect grafts (Figure 5.7A). Figure 5.7B shows the quantitative analysis of the PCR data using densitometry, demonstrating a significantly higher level of IDO expression in the lenti-IDO grafts.  ! "$+!   Figure 5.7 Evaluation of IDO expression in retrieved composite islet grafts. Transplanted composite grafts were retrieved at the time of graft rejection (~2 weeks post-transplant for Lenti-Vect transduced grafts and ~7 weeks post-transplant for Lenti-IDO transduced grafts) and subjected to RT-PCR assay for measurement of human IDO expression. A: RT-PCR analysis of IDO mRNA expression in Lenti-Vect and Lenti-IDO islet grafts. B: Mean±SD ratio of densities of IDO to GAPDH mRNA  (*p<0.001, n=3).  Lenti-Vect graft: islets embedded in mock lentiviral vector-transduced fibroblast populated-CCM. Lenti-IDO graft: islets embedded in IDO gene-transduced fibroblast populated-CCM, ND: not detected.     Considering our findings from immunostaining of grafts, these data suggest that although IDO expression was still detectable by about 7 weeks post-transplant, perhaps the population of Tregs at the graft site was not sufficiently high enough to further prolong graft survival.   0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 A B Lenti-Vect Lenti-IDO IDO  GAPDH Lenti-Vect Lenti-IDO IDO/GAPDH ratio * ! "$"!5.3 DISCUSSION  Re-establishment of the islet ECM is an attractive approach to improve islet survival and function while maintaining healthy islet morphology [111, 203]. The present study shows that employing a novel bioengineered scaffold along with local IDO expression delivered by lentiviral vector significantly prolongs islet allograft survival.        The collagen matrix used in our previous fibroblast populated collagen matrix provided unique features by (i) avoiding direct IDO gene transfer to the islets and potentially compromising insulin production: (ii) applying bystander fibroblasts to maintain the integrity of the matrix while providing essential factors for islet survival; and (iii) using genetically modified fibroblasts as a source of IDO [210]. Fibroblast populated collagen matrix (FPCM) was susceptible, however, to gradual biodegradation and contraction, and moreover IDO gene expression was transient. Therefore, we recently developed a FP-CCM, which has no detectable toxicity for the islets and is more resilient towards contraction and degradation leading to better islet morphology and function [62, 151, 243].        Various viral vectors including adenoviral and lentiviral vectors have been used for gene transduction in islets. The latest generation of adenoviral vectors confer high capacity, reduced inflammation and leads to prolonged gene expression. However, adenoviral vectors are unable to induce life-long transgene expression in vivo, because the viral genome remains episomal in transduced cells [202].  In contrast, lentiviral vectors lead to efficient and more prolonged gene expression in proliferating and non-proliferating cells. Herein, we applied lentiviral gene transduction for local delivery of IDO at the graft site, hoping to generate a more stable non-rejectable islet allograft [202, 256]. The RT-PCR analysis and immunostaining confirmed IDO ! "$#!expression in fibroblasts within the matrix and the majority of incorporated fibroblasts remained viable. Fibroblasts embedded within matrix are barely proliferative and for this reason the number of cells remains relatively constant over time [62, 243].  In order to evaluate the ability of Lenti-IDO fibroblasts to suppress the proliferation of allogeneic immune cells, splenocytes from diabetic C57BL/6 mice, were co-cultured with allogeneic Balb/c islets with or without syngeneic Lenti-IDO fibroblasts. In the present study, splenocytes were isolated from STZ-induced diabetic mice that had already rejected an allograft of Balb/c islets. Therefore, such splenocytes have already been primed with islet alloantigens and are highly alloreactive. Understanding the ability of lentiviral IDO transgene expression to suppress the proliferation of these highly reactive splenocytes after alloantigen-exposure would be crucial. Using flow cytometry, we found that the alloantigen-specific proliferation of splenocytes was markedly suppressed in the presence of Lenti-IDO fibroblasts. Importantly, this effect was almost completely restored by the addition of 1-MT, an IDO inhibitor, indicating that the observed suppressive effect was due to IDO activity.  Our findings showed that the FP-CCM serves as a suitable scaffold for the islets and when transduced with lentiviral vector showed relatively no toxicity for the islets. Immunostaining revealed that when embedded within the matrix, islets maintained insulin and glucagon expressions and were more viable compared to 2D cultured islets. Furthermore, embedding islets within CCM was associated with higher !/$ cell ratio versus 2D cultured islets. Since the higher !/$ cell ratio observed might also be due to differentiation of $ to ! cells, more $ cell death, or ! cell proliferation, these possibilities could be evaluated by lineage tracing of labeled glucagon-producing $ cells, and TUNEL/insulin, TUNEL/glucagon, Ki67/insulin or Ki67/glucagon double staining.  ! "$$!The histology revealed that the FP-CCM perfectly maintains islets healthy structure over time, with minimal cell infiltration and stronger insulin staining in IDO-expressing grafts for up to 7 weeks. The number of intragraft Tregs in IDO-expressing grafts increased significantly within the first 2 weeks post transplantation. Our data revealed that as long as there was sufficient number of FOXP3+ cells at the graft site, islet grafts survived and functioned normally while being protected from T cell infiltration.  To examine levels of IDO produced by Lenti-IDO transduced grafts, harvested islet allografts were evaluated for IDO mRNA expression. Our previous studies showed that Lenti-IDO fibroblast populated cross-linked collagen matrix is capable of expressing IDO for up to 90 days [62]. Herein, the data showed that IDO mRNA was highly expressed in Lenti-IDO allografts, whereas no IDO expression was detected in Lenti-Vect allografts. Since IDO induces FOXP3 expression and plays a crucial role as one of the main downstream mechanisms for certain Tregs. Peripheral expansion of Tregs is known as a downstream mediator of IDO-mediated allograft protection [111, 177, 255]. Our group has also shown that IDO-producing fibroblasts expand the population of antigen-specific Tregs [257]. Therefore, we evaluated the frequency and population of FOXP3+ Tregs at the graft site and in draining lymph nodes of allograft recipients. The population of infiltrating Tregs in Lenti-IDO transduced grafts was significantly higher than in Lenti-Vect transduced grafts at 2 weeks post transplantation however, their number gradually decreased and reached to comparable levels with Lenti-Vect group by week 8 post transplantation. Because IDO expression was still strong at 7 weeks post-transplant in Lenti-IDO grafts, these findings suggest that a decline in IDO expression by fibroblasts is unlikely to be the reason for the rejection of allogeneic islets after ~7 weeks post transplantation. Our previous in vitro studies revealed the stability of IDO expression in ! "$%!composite grafts for up to 90 days (see Chapter 4). Nonetheless, although IDO was noticeably expressed at the graft site, its level might not had been adequate to induce sufficient population of Tregs and protect the allograft against immune rejection for a longer duration. We also suggest that angiogenesis and increased blood flow by 7 weeks post-transplant may dilute the IDO-induced tryptophan metabolites such as kynurenine, which is known to have an anti-proliferative effect on CD4+ and CD8+ cells [216, 232]. Further studies are required to understand the mechanism underlying the lack of longer-term protection of IDO-expressing grafts, such as inducible IDO expression in transplants.        In conclusion, the present study confirms that local intra-graft IDO expression induced in bystander fibroblasts significantly increases islet allograft survival and function. Our data for the first time revealed that the lentiviral transduction method for IDO gene delivery in bystander fibroblasts could significantly improve the duration of islet allograft survival. Having said that, the rejection of allogeneic islets transplanted under the kidney capsule of STZ-induced diabetic mice after about 7 weeks was surprising, as we anticipated Lenti-IDO transduced fibroblasts to generate more stable IDO production and protect transplanted allogeneic islet for much longer duration. Although further studies are clearly needed to delineate the mechanism of loss of long-term IDO-mediated protection, these studies raise the possibility that local IDO expression could be used as an adjunct with other low dose immunosuppression for prevention of islet allograft rejection. Considering the potential benefits of IDO in cell replacement therapies, further strategies should be evaluated to provide sufficient IDO expression or IDO-mediated FOXP3+ Tregs at the graft site to prevent allograft rejection.   ! "$&!          Chapter 6:  DISCUSSION AND CONCLUSIONS            ! "$'!6.1 SUMMARY AND DISCUSSION  Since the discovery of the immunoregulatory characteristics of IDO, this enzyme has been shown to have far-ranging roles in mammalian pregnancy [222], autoimmune diseases [216], inflammation [258], allergy [172], neoplasia [259] and most recently inhibition of alloimmune responses in transplantation [172, 210, 219]. As mentioned before, IDO catalyzes the rate-limiting step of tryptophan metabolism along the kynurenine pathway that leads to tryptophan deficiency and production of kynurenine metabolites. The consequence of IDO activity is induction of T cell apoptosis, restraining T cell proliferation, and promoting Tregs expansion [162, 193]. Under normal conditions, IDO is expressed at low basal levels by plasmocytoid DCs; however, it can be induced rapidly by IFN# via the JAK/STAT1 pathway [260] (Figure 6.1). There is compelling evidence that DCs play an important role in modulation of the immune response and tolerance. Previously, Grohmann et al [214] found that IFN! fails to potentiate the tolerogenic properties of DCs from prediabetic female NOD mice and suggested that such a phenomenon could play a role in the development of diabetes and impaired immunotolerance in these mice [214]. In the present study, we examined the ability of NOD dermal fibroblasts to increase IDO expression and tryptophan catabolism in response to IFN#. IFN# is a cytokine that can have both stimulatory and inhibitory effects on the immune system. In the acquired immune response, IFN# is an important effector molecule released by antigen-specific CD4+ and CD8+ T cells [214, 260]. In the course of an immune response to a foreign antigen, IFN# induced-IDO expression can modulate T cell homeostasis [158]. IFN# is a key factor playing an important role in regulating the activation of bystander and accessory cells along with autoreactive lymphocytes that are involved in the autoimmune response: IDO-expressing DCs, that are activated by IFN#-expressing autoreactive T cells, maintain the ! "$(!peripheral tolerance in part through suppression of T cell proliferation and induction of apoptosis [158].  Fibroblasts are important components of islet ECM. They not only provide structural support and regulate the tensile strength of ECM but also maintain the integrity of islets by producing collagen fibers and other important components of ECM such as fibronectins [217]. Several studies have demonstrated far-ranging roles for fibroblasts suggesting them to be involved in modulation of the immune response [224, 225].   It has been well documented that fibroblasts are non-professional APCs [224, 225]. In addition, Jones et al have demonstrated that dermal fibroblasts are potent immunoregulatory cells and are able to inhibit T cell proliferation by inducing G0/G1 cell cycle arrest, and decreasing the secretion of proinflammatory cytokines [261]. Hannifa et al., showed that fibroblast-mediated inhibition of alloreactive T cells is critically dependent on IFN# released from activated T cells. IFN#-induced IDO expression in fibroblasts, and perhaps enhanced tryptophan catabolism, are at least partly responsible for suppression of T cell proliferation [262].  Keeping in mind that fibroblasts play a crucial role in modulation of the islet environment and considering the defective IDO expression in DCs of early prediabetic NOD mice in response to IFN# treatment, we asked whether this same phenomenon would be observed in NOD dermal fibroblasts. While confirming the data obtained by Grohmann et al. in DCs, our data revealed that NOD dermal fibroblasts fail to express IDO and catabolize tryptophan independent of gender or status of diabetes progression. Our mechanistic investigation of the IFN# signaling pathway revealed that, in contrary to the defective IFN#-mediated IDO expression pathway in NOD mouse fibroblasts, other IFN# mediated-pathways independent of JAK/STAT1 including IFN#-induced MHC I expression (mediated via both JAK/STAT1 and ! "$)!canonical NF-%B pathways) and IFN#-mediated type 1 collagen repression (facilitated through the activity of C/EBP!) were operative and the observed defect in this pathway did not interfere with other aspects of IFN-#-mediated gene expressions and function. Further investigation revealed that NOD dermal fibroblasts efficiently express IDO when transduced with an adenoviral vector carrying the IDO gene, indicating that the defect in this pathway must occur proximal to mRNA and protein synthesis. We showed that defective STAT1 phosphorylation in NOD dermal fibroblasts impairs the IFN#-mediated JAK/STAT1 pathway resulting in failure of these cells to express IDO in response to this cytokine. Finally, we showed that an IFN#-independent IDO expression pathway (i.e. LPS-mediated-JNK) is operative in NOD dermal fibroblasts. This finding is of importance since it represents the proficiency of these cells to modulate the immune response when encountering bacterial infections, where LPS could play an important role in induction of IDO expression. Additionally, we suggest that if the same phenomenon is observed in pancreatic fibroblasts, islets would be less protected from autoimmune attack due to defective IDO expression of fibroblasts in response to IFN#. Further studies are required to investigate this hypothesis.  Islet transplantation has emerged in recent years as a highly promising therapy for type 1 diabetes that replaces ! cells and enables proper control of blood glucose, decreased risk of complications and improved quality of life in graft recipients. In the current study, we sought to implement two strategies to improve islet transplantation outcome: i) Mimicking the native islet microenvironment and replacing islet ECM, lost during isolation process; and ii) Generating a local immunomodulated microenvironment using stable IDO-expressing fibroblasts to avoid side effects associated with systemic immunosuppressive regimens.  ! "$*!In our first strategy, we generated and applied a novel bioengineered cross-linked collagen-glycosaminoglycan hydrogel scaffold. Our in vivo data show that embedding islets in a fibroblast-populated cross-linked collagen matrix (FP-CCM) significantly improves insulin and glucagon expression and maintains the ratio of !/$ cells when compared to 2D cultured islets. A 100-day follow up on syngeneic transplantation of islets embedded in FP-CCM, confirmed that islets remained viable and functional and maintained normal morphology within this matrix. Our in vitro data also demonstrate that the FP-CCM significantly improves islet viability when compared to either islets alone or co-cultured with fibroblasts in collagen matrix (fibroblast populated collagen matrix; FPCM), 30 days post culture. We also showed that incorporating islets within the FP-CCM maintains insulin and glucagon production in vitro. Compared to islets transplanted alone, islet survival was improved when embedded prior to transplant within FP-CCM, as indicated by the lack of cleaved caspase-3 as seen by immunostaining. Furthermore, our proposed matrix significantly preserved the insulin secretory function of incorporated islets when compared to islets alone over a 30 day-assessment period. The finding of this study also ruled out the possibility of fibrosis in FP-CCM as the proliferation of fibroblasts significantly decreased when embedded within the matrix.   Our proposed novel islet scaffold is superior to our previously generated FPCM for several reasons. Fibroblasts incorporated within free-floating collagen gels tend to contract. The FP-CCM not only reduces the proliferation of fibroblasts but also presents improved tensile strength and thus is more resilient towards contraction. In such a matrix, islets maintain healthy morphology and therefore function. Furthermore, the FP-CM is more likely to undergo gradual biodegradation; fragmented collagen fibrils can induce immune and fibrotic responses. However, ! "%+!the FP-CCM is less prone to enzymatic degradation because of glutaraldehyde-collagen and PVA-borate cross-links.  Our second strategy was to apply stable IDO-producing fibroblasts to incorporate within the CCM along with islets. For this reason we used a viral vector delivery system to genetically modify bystander fibroblasts to express IDO gene. Development of an efficient and safe gene transduction method that is amenable for clinical applications is crucial. An optimal gene transduction system should maintain long-term gene expression levels, without any toxic or inflammatory side effects.  The use of a non-integrating vector like adenovirus has the advantage of lower risk of oncogene activation. However, several studies have shown a gradual decrease in transgene expression using this vector. Another disadvantage of adenoviral vectors is their effectual interactions with APCs resulting in activation of innate and adaptive immune responses against the expressed protein and the expressing cell. Integrating vectors such as lentiviral vectors offer the advantage of efficient gene transduction in both dividing and non-dividing cells. The most important safety concern related to all integrating vectors is the risk of oncogene activation, tumor-suppressor gene inactivation or insertional mutagenesis [257]. Our group has previously applied adenoviral transduction for generation of IDO-producing fibroblasts. Using Ade-IDO fibroblasts embedded in collagen matrix, we were able to extend islet allograft survival to an average of ~42 days. In the present study we applied a lentiviral vector for IDO gene transduction to generate more stable IDO expression (Lenti-IDO transduced fibroblasts). One of the advantages of our proposed system is avoiding direct islet gene delivery but transducing IDO expression in bystander fibroblasts. When incorporating Lenti-IDO fibroblasts in CCM, our data revealed that IDO transgene expression was stable for at least 90 days.  Furthermore, local IDO expression delivered by lentiviral vector did not jeopardize islet survival or function.  ! "%"!IDO not only has been recognized as a part of the innate immune defense against certain pathogens, but also has been shown to regulate adaptive T cell immunity [112]. The recognition of IDO as a local immunoregulatory factor in mammalian pregnancy was a critical step in further defining the importance of IDO in modulation of the immune response. Since then, numerous studies have investigated the role of IDO in immunological aspects of physiologic and pathologic conditions like pregnancy, cancer, infections, autoimmune diseases and transplantation. IDO activity links two arms of innate and adaptive immunity to create local immunosuppression, and to encourage systemic tolerance by activating Tregs.  As mentioned before, IDO induces activation of GCN2 in CD8+ T cells leading to cell-cycle arrest and functional anergy [209]. IDO-induced activation of GCN2 in CD4+ cells, together with kynurenine metabolites, induces de novo differentiation of FOXP3+ Tregs from uncommitted CD4+ T cells while promoting effector function in mature FOXP3+ Tregs. Tregs are also able to generate tolerogeneic DCs through the induction of IDO expression by CTLA-4 interactions and, in turn, tolerogeneic DCs can contribute to the spreading of immunosuppression via a phenomenon known as “infectious tolerance” [163]. Through this phenomenon, IDO establishes a tolerogenic phenotype in DCs and expands the population of Tregs [263].   The present study clearly supports the role of IDO in protection of islet allograft survival. Our data revealed that when our novel bioengineered islet graft is equipped with long-lasting IDO-producing cells, islet allografts were protected and animals remained euglycemic for an average of ~51 days versus control groups, who return to hyperglycemia within less than two weeks following transplantation.  ! "%#!   Figure 6.1 Proposed model of IDO-mediated infectious tolerance. Following generation of Tregs by IDO-expressing DCs, they might in turn create other tolerogenic DCs through the induction of IDO expression by CTLA-4 interactions: a mechanism that explains a self-amplifying regulatory network, known as infectious tolerance.     As mentioned before, although this result represents a significant prolongation of islet allograft survival, we anticipated a longer protection of allograft using the lentiviral gene transduction method. Our in vitro data clearly confirmed IDO expression in Lenti-IDO transduced fibroblasts embedded within the cross-linked collagen matrix for up to 90 days. Our in vivo data revealed Regulatory T cell Naïve T cell IDO- competent  Naï ve  DC IDO+ Regulatory  DC CTLA-4 IDO ! "%$!that at the time of graft rejection in IDO-islet graft recipients, human IDO expression was still detectable at the graft site. Nevertheless, immunofluorescence staining showed that the population of FOXP3+ cells in IDO-expressing grafts was similar to that of non-IDO expressing grafts at the time of graft failure. One explanation to this observation could be that IDO expression at the graft site might not have been sufficient to maintain the FOXP3+ cells population, which are downstream effector cells of IDO-mediated immunomodulation.   6.2 SIGNIFICANCE Type 1 diabetes remains an incurable disease with devastating complications despite insulin therapy. Proper management or a cure for diabetes will require a deeper understanding of the pathogenesis of the disease. Although ! cell replacement by islet transplantation has a great potential for treating patients with type 1 diabetes, the lifelong systemic use of immunosuppressive medications following transplantation remains a major challenge: the non-specific and systemic nature of these medications may expose the patients to risk of serious side-effects, including !-cell toxicity, hyperlipidemia, nephrotoxicity, increased risk of both infection and developing certain malignant diseases e.g., lymphomas.  In this project we first found a defect in dermal fibroblasts in an animal model of human type 1 diabetes that could open new avenues for better understanding the mechanisms underlying progression of type 1 diabetes.  Furthermore, we sought a desirable solution by which viable islet cells could be transplanted to diabetic animal models without the need for anti-rejection medications. The composite graft used in this work is a novel bioengineered liquid scaffold that promotes the viability and function of the islets. By this approach, we aim to address one of the ! "%%!major concerns facing clinical islet transplantation, by replacing insulin therapy with anti-rejection medications while improving islet graft function and survival. Based on these results we hypothesize that if there is sufficient IDO expression at the graft site, islet allografts could be protected against allo-rejection. Therefore, we believe that these findings have the potential to lead to development of a new therapeutic option through which type 1 diabetic patients could receive islet transplants with relatively lower doses of toxic immunosuppressive drugs commonly used after transplantation surgeries.  6.3 FUTURE STUDIES To address the hypothesis that defective IFN#-induced IDO expression from pancreatic fibroblasts in the ECM of islets could contribute to susceptibility of islet cells to autoimmune attack, we must understand whether the same phenomenon occurs in NOD pancreatic fibroblasts or not. If it proves to be true, targeted IDO gene therapy to the pancreas could be a feasible approach to partially prevent or decelerate the progression of type 1 diabetes. Clinical translation of this hypothesis requires isolation of diabetic human dermal and pancreatic fibroblasts for evaluation of IDO expression. Our data suggest that defective IFN#-induced IDO expression in NOD dermal fibroblasts is independent of gender or diabetes stage. If it proves to be true for human fibroblasts as well, one could hypothesize that IFN#-induced IDO expression could be applied as a marker of early diagnosis of diabetes in susceptible individuals with at-risk genetic backgrounds.  We also demonstrate the successful utilization of a novel bioengineered matrix equipped with a local immunosuppressive enzyme, IDO, for islet transplantation. This matrix can ! "%&!substitute for the natural ECM of the islets that is lost during isolation processes, and offers proper mechanical strength, while local IDO expression delivered by bystander fibroblasts significantly prolongs allograft survival. Considering that both allo- and autoimmunity are involved in allograft rejection [209], the streptozotocin-induced model of diabetes is not an ideal model as it does not address the contribution of recurrent autoimmunity in graft rejection. Therefore, for clinical translation of our approach, this novel IDO-expressing islet graft should be examined in an animal model in which autoimmune mechanisms contribute to graft rejection: the NOD mouse. This model is a well-studied animal model of type 1 diabetes that exhibits all the characteristics of typical human autoimmune diabetes including macroscopic, histopathological and biochemical alterations along with presenting acute and chronic symptoms [210]. As such, the NOD mouse strain would be an appropriate model to investigate the effectiveness of the proposed engraftment of allogeneic pancreatic islets against both allo- and autoimmune rejection. Additionally human and murine immune systems marginally differ in their development and activation both in the innate and adaptive responses [264]. Our proposed approach could be also tested in a humanized mouse model using immune-deficient NOD mice that allow engraftment of human tissue and hematopoietic cells, enabling the study of human islet transplantation in the presence of a functioning human immune system [265]. Further studies could also include generating more efficient gene transduction methods and specifically, to make this model more translatable to the clinic, non-viral stable gene-transfer methods to induce IDO in co-transplanted fibroblasts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`,K!!"#$*)8#$),$@AB%C$&2$&'8#($(*.2'18.2(.(&)2K!/H!DE1#*&:#2(.8$F#3&5&2#,!#+""K!aC1!bH/91>6/0I!?=!Y>/0/6C!N?;4@</:!!"+,! Z1/>K![,N,K!N,!N:91;0/GZ121>K!:H2!O,!Y?HH1>GZ1/>K!G(#:$5#88$.11*).5"#'$,)*$3&./#(#'H$()>.*3'$/#(.$5#88$*#18.5#:#2(4![1H?@1!J12K!#+"",!,T*UA!D,!'",!"",! Y?/0:>2K!N,K!9.25*#.(&5$&'8#($.6()&::62&(+4!W>1661!J12K!#+"#,!*(T"#!D!#UA!D,!1'$'G&+,!"#,! _14H5K!Z,N,K!Z,M,!c:F;/H6?HK!:H2!J,L,!N>:/5K!D2(#*)I&*6'$&2,#5(&)2$.23$(+1#$<$3&./#(#'$:#88&(6'H$'+'(#:.(&5$*#I&#>$.23$:#(.B.2.8+'&'$),$)/'#*I.(&)2.8$:)8#568.*$'(63&#'4!Y@dK!#+"",!,*"A!D,!2$&,!"$,! eH/DK!J,K!10!:;,K!-&#(.*+$&2(#*I#2(&)2$&2$&2,.25+$.23$8.(#*$'&J2'$),$/#(.B5#88$.6()&::62&(+4!S!LH5;!^!J12K!#+"+,!,-,T#+UA!D,!"*++G),!"%,! eH/DK!J,K!O,J,!P/>0:H1HK!:H2!\,e,!3Q1><;?@K!=2,.2($,##3&2J$.23$("#$*&';$),$(+1#$<$3&./#(#'4!3@!^!N;/H!S40>K!#+"+,!.(T&UA!D,!"&+'6G"&"$6,!"&,! Y/6.C?==K!],\,K!K"#:);&2#$:#3&.(#3$:)368.(&)2$),$.6()&::62&(+$&2$(+1#$<$3&./#(#'$/H!9.(")8)J+$.23$L./)*.()*+$F#3&5&2#,!#+"#K!aC1!bH/91>6/0I!?=!Y>/0/6C!N?;4@</:,!"',! G(.23.*3'$),$:#3&5.8$5.*#$&2$3&./#(#'BBMN<O4!M/:<1016!N:>1K!#+"$,!,-!$%&&'!(A!D,!O""G'',!"(,! c?<1>06?HK!c,W,K!='8#($(*.2'18.2(.(&)2$.'$.$(*#.(:#2($,)*$3&./#(#'$B$.$>)*;$&2$1*)J*#''4!S!LH5;!^!J12K!#++%,!,+/T(UA!D,!'*%G(+&,!"),! S:>1H2>:HK!W,K!L,!L601;;:K!:H2!O,!X?4>;:H?6K!=::62)8)J+$),$(+1#$<$3&./#(#'4!fd@K!#++&,!.)T)UA!D,!&%(G&',!"*,! J?>>:HK!J,W,K![,O,!g@1HHK!:H2!J,!W/10>?D:?;?K!=::62)8)J+$.23$J#2#(&5'$),$(+1#$<$3&./#(#'4!J0!O/H:/!^!J12K!#++),!#+T%UA!D,!$"%G#(,!#+,! N?DD/101>6K!e,a,K!10!:;,K!-#:)2'(*.(&)2$),$&'8#(B.6()*#.5(&I#$K-P$!$5#88'$&2$&2'68&(&5$8#'&)2'$,*):$*#5#2($)2'#($.23$8)2JB(#*:$(+1#$<$3&./#(#'$1.(&#2('4!^!LVD!J12K!#+"#,!"/.T"UA!D,!&"G'+,!! "%(!#",! Z/;;.?VK!3,K!10!:;,K!72.8+'&'$),$&'8#($&2,8.::.(&)2$&2$"6:.2$(+1#$<$3&./#(#'4!N;/H!LVD!R@@4H?;K!#++*,!(++T#UA!D,!"($G)",!##,! RHh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e:0iK!^,M,K!N,!Y1H?/60K!:H2!M,!J:0C/6K!!$"#81#*$5#88$'6/'#('$&2$&2'68&2B3#1#23#2($3&./#(#'4!O./1H.1K!"**&,!"-)T&#"%UA!D,!"")&G),!#*,! Z?H5K!X,O,!:H2!N,3,!^:H1F:IK!^>,K!!"#$*)8#$),$K-C$.23$K-P$!$5#88'$&2$(+1#$=$3&./#(#'$&2$("#$0U-$:)6'#4!c16!R@@4H?;K!"**(,!(*)T&UA!D,!$#(G$#,!$+,! _??HK!^,Z,K!\,O,!^4HK!:H2!W,!O:H0:@:>/:K!K#8868.*$.23$:)8#568.*$:#5".2&':'$,)*$("#$&2&(&.(&)2$.23$1*)J*#''&)2$),$/#(.$5#88$3#'(*65(&)2$*#'68(&2J$,*):$("#$5)88./)*.(&)2$/#(>##2$:.5*)1".J#'$.23$!$5#88'4!340?/@@4H/0IK!"**),!"#T#UA!D,!"+*G##,!$",! ]11K!e,b,K!10!:;,K!9*#,#*#2(&.8$&2,&8(*.(&)2$),$:.5*)1".J#'$36*&2J$#.*8+$'(.J#'$),$&2'68&(&'$&2$3&./#(#'B1*)2#$@@$*.('4!M/:<1016K!"*)),!,#T)UA!D,!"+&$G),!$#,! _??HK!^,Z,!:H2!\,O,!^4HK!K#8868.*$.23$:)8#568.*$1.(")J#2&5$:#5".2&':'$),$&2'68&2B3#1#23#2($3&./#(#'$:#88&(6'4!3HH!S!_!3.:2!O./K!#++",!.")A!D,!#++G"",!$$,! W1:>;G_:=1K!J,K!10!:;,K!9.25*#.(&5$&'8#('$623#*$.((.5;H$5#8868.*$.23$:)8#568.*$#,,#5()*'4!N4>>!WC:>@!M16K!#++(,!(,T(UA!D,!(%*G'+,!$%,! L/d/!e:F:6:Q/K!c,[,![/;;K!:H2![,O,!L/61H<:>0CK!!+1#$=$-&./#(#'$F#88&(6'K!/H!D23)5*&2#$.23$U*J.2$G1#5&,&5$76()&::62&(+,!"***K!]:H216!Y/?6./1H.1,!$&,! Z:H5K!`,K!10!:;,K!@AB%C$9.(">.+$&2$='8#($!*.2'18.2(.(&)2$.23$/#(.BK#88$T#18.5#:#2($!"#*.1&#'4!^!a>:H6D;:H0K!#+"",!"/((A!D,!%")*+#,!$',! e>14F1;K!\,a,K!10!:;,K!K):1.*&2J$("#$*#8.(&I#$*)8#$),$1#*,)*&2VJ*.2W+:#$I#*'6'$X.'VX.'$8&J.23$5+()()E&5$1.(">.+'$&2$K-PY$!$5#88B:#3&.(#3$&2'68&2B3#1#23#2($3&./#(#'$:#88&(6'4!^!R@@4H?;K!"***,!(-,T)UA!D,!%$$&G%",!$(,! W1:Q@:HK!J,K!=::62)8)J&5.8$1.(">.+'$()$/#(.B5#88$3.:.J#$&2$!+1#$<$3&./#(#'4!M/:<10!J12K!#+"$,!,/T#UA!D,!"%(G&%,!$),! ]/H2K!e,K!J,\,!\4CHK!:H2!J,!X;?260>?@Ga4;;<1>5K!=::62)8)J+$&2$("#$58&2&5$*#I&#>$'#*&#'Z$,)56'$)2$(+1#$<$3&./#(#'$.23$I&*6'#'H$("#$&22.(#$&::62#$*#'1)2'#$()$#2(#*)I&*6'#'$.23$&('$1)''&/8#$*)8#$&2$*#J68.(&2J$(+1#$<$3&./#(#'4!N;/H!LVD!R@@4H?;K!#+"#,!(-)T"UA!D,!$+G),!$*,! j/15;1>K!3,[,!:H2![,a,!S1D?@K!9*#3&5(&)2$.23$1.(")J#2#'&'$&2$(+1#$<$3&./#(#'4!R@@4H/0IK!#+"+,!,"T%UA!D,!%')G(),!%+,! ]/K!Z,K!10!:;,K!K))1#*.(&)2$),$&2I.*&.2($0[!$5#88'$.23$K-CYK-M\Y$!$*#J68.()*+$5#88'$&2$1*#I#2(&)2$),$.6()&::62#$3&./#(#'$&2$2)2B)/#'#$3&./#(&5$:&5#$(*#.(#3$>&("$.81".BJ.8.5()'+85#*.:&3#4!3.0:!Y/?.C/@!Y/?DCI6!O/H!TOC:H5C:/UK!#++),!*/T&UA!D,!$)"G*+,!! "%)!%",! ^:/;F:;:K!W,K!10!:;,K!71)1()'&'$),$K-CY$K-M\]"&J"^$!$5#88'$&2$(+1#$<$3&./#(#'$:.+$/#$1.*(&.88+$:#3&.(#3$/+$=LBM$3#1*&I.(&)24!W]?O!gH1K!#++*,!*T)UA!D,!1'&#(,!%#,! J:6?HK!M,!:H2!X,!W?F>/1K!K)2(*)8$),$&::62#$1.(")8)J+$/+$*#J68.()*+$!$5#88'4!N4>>!gD/H!R@@4H?;K!"**),!(/T'UA!D,!'%*G&&,!%$,! c?@:5H:H/K!O,K!!+1#$<$!$"#81#*$.23$(+1#$M$!$"#81#*$5#88'H$,625(&)2'?$*#J68.(&)2$.23$*)8#$&2$1*)(#5(&)2$.23$3&'#.'#4!RH0!^!N;/H!]:<!c16K!"**",!"(T#UA!D,!"&#G),!%%,! M?H5K!N,!:H2!c,3,!X;:91;;K!K#88$,.(#$3#5&'&)2H$!B"#81#*$<$.23$M$'6/'#('$&2$&::62#$*#'1)2'#'4!3>0C>/0/6!c16K!#+++,!"T$UA!D,!"(*G")),!%&,! [?>K!M,g,K!S,c,!c?61K!:H2!S,O,![>11H6D:HK!!%<B!%MH$.$1*)5*6'(#.2$1.*.3&J:4!S:0!R@@4H?;K!#++$,!*T'UA!D,!&+$G&,!%',! Z/;6?HK!O,Y,K!10!:;,K!DE(*#:#$!"<$/&.'$),$&2I.*&.2($_.81".MC`.81".a$!$5#88'$&2$(+1#$<$3&./#(#'4!S:04>1K!"**),!,.(T'''$UA!D,!"((G)",!%(,! L/i/>/QK!M,],K!J,],!N?;;/K!:H2!X,!g>0/6K!!"#$*)8#$),$&2,8.::.(&)2$&2$&2'68&(&'$.23$/#(.B5#88$8)''$&2$(+1#$<$3&./#(#'4!S:0!c19!LH2?.>/H?;K!#++*,!+T%UA!D,!#"*G#',!%),! 3Q:064Q:K!\,K!10!:;,K!7$5.'#$),$,68:&2.2($(+1#$<$3&./#(#'$>&("$5)E'.5;&#$@C$I&*6'$&2,#5(&)2$3&.J2)'#3$/+$#8#I.(#3$'#*6:$8#I#8'$),$2#6(*.8&W&2J$.2(&/)3+4!M/:<1016!c16!N;/H!W>:.0K!#++*,!)*T$UA!D,!1&+G#,!%*,! 9:H!Y1;;1K!a,],K!e,a,!N?DD/101>6K!:H2!J,[,!9?H!\1>>:0CK!!+1#$<$3&./#(#'H$#(&)8)J+?$&::62)8)J+?$.23$("#*.1#6(&5$'(*.(#J&#'4!WCI6/?;!c19K!#+"",!.(T"UA!D,!(*G""),!&+,! J:>/100:K!L,P,K!10!:;,K!L)>$&25&3#25#$),$'1)2(.2#)6'$(+1#$<$3&./#(#'$&2$2)2B)/#'#$3&./#(&5$:&5#$*.&'#3$)2$J86(#2B,*##$3&#('$&'$.'')5&.(#3$>&("$5".2J#'$&2$("#$&2(#'(&2.8$:&5*)/&):#4!W]?O!gH1K!#+"$,!)T""UA!D,!1()')(,!&",! X4H2:K!M,W,K!10!:;,K!S86(#2B,*##$3&#($1*#I#2('$3&./#(#'$&2$0U-$:&5#4!M/:<1016!J10:<!c16!c19K!"***,!(+T&UA!D,!$#$G(,!&#,! O9?>1HK!Y,J,K!10!:;,K!G&J2&,&5.2($I&(.:&2$-$3#,&5&#25+$&2$+)6("$>&("$(+1#$<$3&./#(#'$:#88&(6'4!^!W12/:0>K!#++*,!(+*T"UA!D,!"$#G%,!&$,! N??D1>K!^,M,K!10!:;,K!=2"#*&(#3$I.*&.(&)2$&2$I&(.:&2$-$J#2#'$&'$.'')5&.(#3$>&("$1*#3&'1)'&(&)2$()$.6()&::62#$3&'#.'#$(+1#$<$3&./#(#'4!M/:<1016K!#+"",!-/T&UA!D,!"'#%G$",!&%,! a4>4H1HK!^,3,K!10!:;,K!7'')5&.(&)2$.2.8+'&'$),$("#$7=TD$.23$&2'68&2$J#2#'$&2$X&22&'"$(+1#$<$3&./#(&5$1.(&#2('4!R@@4H?51H10/.6K!#++',!+)T&G'UA!D,!$$"G),!&&,! Z/;2/HK!c,O,K!10!:;,K!b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c!N4>>!M/:<!c1DK!#+"+,!(/T'UA!D,!&+'G"",!&),! aC?@D6?HK!M,J,K!10!:;,K!T#365#3$1*)J*#''&)2$),$3&./#(&5$:&5*)I.'568.*$5):18&5.(&)2'$>&("$&'8#($5#88$(*.2'18.2(.(&)2$5):1.*#3$>&("$&2(#2'&I#$:#3&5.8$("#*.1+4!a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a:<:0:<:1/K!3,K!10!:;,K!D:/#33&2J$&'8#($&2$.$8&Q6&3$'5.,,)83$&25*#.'#'$&'8#($I&./&8&(+$.23$,625(&)24!N:H!^!M/:<1016K!#+"$,!,#T"UA!D,!#(G$&,!'$,! cI:HK!L,3,K!10!:;,K!K8&2&5.8$)6(5):#'$.23$&2'68&2$'#5*#(&)2$.,(#*$&'8#($(*.2'18.2(.(&)2$>&("$("#$D3:)2()2$1*)()5)84!M/:<1016K!#++",!+/T%UA!D,!("+G*,!'%,! J:064@?0?K!O,K!='8#($5#88$(*.2'18.2(.(&)2$,)*$!+1#$<$3&./#(#'4!^!M/:<1016K!#+"+,!"T"UA!D,!"'G##,!'&,! c116K!M,3,!:H2!^,N,!3;.?;:2?K!72&:.8$:)3#8'$),$3&./#(#'$:#88&(6'4!M/:<10!J12K!#++&,!""T%UA!D,!$&*G(+,!'',! Y>1i:>K!P,K!10!:;,K!@#+)23$("#$")*:)2#H$&2'68&2$.'$.2$.6()&::62#$(.*J#($&2$(+1#$<$3&./#(#'4!LH2?.>!c19K!#+"",!,"T&UA!D,!'#$G'*,!'(,! [>1/H1>K!M,],K!10!:;,K!!"#$!$5#88$:.*;#*$T!d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c,3,!O0?Q16K!:H2!^,L,![4H0?HK!='8#($(*.2'18.2(.(&)2H$,.5()*'$&2$'")*(B(#*:$&'8#($'6*I&I.84!3>.C!R@@4H?;!aC1>!LVD!TZ:>6iUK!#+"",!+.T'UA!D,!%#"G*,!(#,! e/HK!a,K!10!:;,K!='8#($&')8.(&)2$.23$(*.2'18.2(.(&)2$)6(5):#'$),$1.25*#.'$1*#'#*I#3$>&("$e2&I#*'&(+$),$f&'5)2'&2$')86(&)2$I#*'6'$(>)B8.+#*$:#(")3$6'&2J$1*#)E+J#2.(#3$1#*,86)*)5.*/)24!a>:H6D;:H0:0/?HK!#++',!)"T"+UA!D,!"#)'G*+,!($,! J:064@?0?K!O,!:H2!_,!e4>?2:K!9#*,86)*)5.*/)2$,)*$)*J.2$1*#'#*I.(&)2$/#,)*#$(*.2'18.2(.(&)24!a>:H6D;:H0:0/?HK!#++#,!#*T"#UA!D,!")+%G*,!(%,! O:F:2:K!a,K!10!:;,K!=:1*)I#3$&'8#($+&#83$.23$,625(&)2$>&("$365(.8$&2g#5(&)2$),$e2&I#*'&(+$),$f&'5)2'&2$')86(&)2$/#,)*#$1.25*#.'$1*#'#*I.(&)24!a>:H6D;:H0:0/?HK!#++$,!#+T"#UA!D,!"*'&G*,!(&,! c/.?>2/K!N,K!10!:;,K!%6:.2$&'8#($&')8.(&)2$.23$16*&,&5.(&)2$,*):$1#3&.(*&5B.J#$3)2)*'4!a>:H6D;:H0!W>?.K!"**",!",T"!W0!"UA!D,!()$G%,!(',! J:064@?0?K!O,K!10!:;,K!=:1*)I#:#2($),$1.25*#.(&5$&'8#($5#88$&')8.(&)2$,)*$(*.2'18.2(.(&)24!W>?.!TY:I;!bH/9!J12!N1H0UK!#++(,!"/T%UA!D,!$&(G'#,!((,! \1>/H5K!Y,^,K!10!:;,K!!"#$#I.86.(&)2$),$2#6(*.8$3#2'&(+$'#1.*.(&)2$6(&8&W&2J$X&5)88B')3&6:$3&.(*&W).(#$.23$0+5)3#2W$.23$5#2(*&,6J.8$#86(*&.(&)2$&2$("#$16*&,&5.(&)2$),$/)I&2#$.23$5.2&2#$&'8#($1*#1.*.(&)2'4!\?>@!J10:<!c16!O4DD;K!"**+,!"+A!D,!&(G'$,!(),! Y:;;/H51>K!Z,X,!:H2!W,L,!]:.IK!!*.2'18.2(.(&)2$),$&2(.5($1.25*#.(&5$&'8#('$&2$*.('4!O4>51>IK!"*(#,!#"T#UA!D,!"(&G)',!(*,! c1.Q:>2K!N,c,!:H2!N,X,!Y:>Q1>K!!*.2'18.2(.(&)2$),$&')8.(#3$1.25*#.(&5$&'8#('$.5*)''$'(*)2J$.23$>#.;$"&'()5):1.(&/&8&(+$/.**&#*'4!a>:H6D;:H0!W>?.K!"*($,!+T"UA!D,!('"G$,!)+,! OC:D/>?K!3,J,K!10!:;,K!='8#($(*.2'18.2(.(&)2$&2$'#I#2$1.(&#2('$>&("$(+1#$<$3&./#(#'$:#88&(6'$6'&2J$.$J865)5)*(&5)&3B,*##$&::62)'611*#''&I#$*#J&:#24!S!LH5;!^!J12K!#+++,!,*,T%UA!D,!#$+G),!! "&+!)",! cI:HK!L,3,K!10!:;,K!X&I#B+#.*$,)88)>B61$.,(#*$58&2&5.8$&'8#($(*.2'18.2(.(&)24!M/:<1016K!#++&,!+*T(UA!D,!#+'+G*,!)#,! ^:@/?;Q?F6Q/K!c,J,K!10!:;,K!='8#($(*.2'18.2(.(&)2$&2$(+1#$=$3&./#(#'$:#88&(6'4!_:;1!^!Y/?;!J12K!#+"#,!)+T"UA!D,!$(G%$,!)$,! ='8#($K#88$!*.2'18.2($=:.J#',!#+"$!!-./012!#+"$!"$!34546078!39:/;:<;1!=>?@A!C00DAEEFFF,2/:<1016>161:>.C,?>5,!)%,! J.N:;;K!J,!:H2!3,J,!OC:D/>?K!e13.(#$)2$&'8#($(*.2'18.2(.(&)24!N?;2!OD>/H5!\:><!W1>6D1.0!J12K!#+"#,!"T(UA!D,!:++()#$,!)&,! Y:.CK!X,\,!:H2!M,Y,!3@?6K!%6B<H$F.g)*$"&'()5):1.(&/&8&(+$8)56'$&2$:.24!O./1H.1K!"*'(,!(+-T$()"UA!D,!"&+'G),!)',! NC/H1HK!^,!:H2!c,\,!Y4.Q;1IK!!*.2'18.2(.(&)2$&::62)8)J+H$')8&3$)*J.2$.23$/)2#$:.**)>4!^!3;;1>5I!N;/H!R@@4H?;K!#+"+,!("+T#!O4DD;!#UA!D,!O$#%G$&,!)(,! J?>1:4K!3,K!10!:;,K!D,,#5()*$:#5".2&':'$),$*#g#5(&)24!N?;2!OD>/H5!\:><!W1>6D1.0!J12K!#+"$,!,T""U,!)),! K)88./)*.(&I#$='8#($!*.2'18.2($T#J&'(*+$]K=!T^?$A("$7226.8$T#1)*($#+"+!!-./012!#+"$!"$!34546078!39:/;:<;1!=>?@A!C00DAEEFFF,./0>15/60>I,?>5E!!)*,! J:064@?0?K!R,K!10!:;,K!=:1*)I#:#2($&2$&'8#($+&#83$,*):$)/#'#$3)2)*'$,)*$"6:.2$&'8#($(*.2'18.2('4!a>:H6D;:H0:0/?HK!#++%,!#)T'UA!D,!))+G&,!*+,! RC@K!O,\,K!10!:;,K!D,,#5($),$3)2)*$.J#$)2$,625(&)2$),$&')8.(#3$"6:.2$&'8#('4!M/:<1016K!#++',!++T&UA!D,!"$'"G),!*",! ]/4K!`,K!10!:;,K!72.8+'&'$),$3)2)*B$.23$&')8.(&)2B*#8.(#3$I.*&./8#'$,*):$2)2B"#.*(B/#.(&2J$3)2)*'$]0%@-'^$6'&2J$("#$[+)()$&'8#($&')8.(&)2$:#(")34!N1;;!a>:H6D;:H0K!#++),!(#T'UA!D,!'%*G&',!*#,! W?001>K!e,^,K!10!:;,K!-#.("$.23$-+',625(&)2$),$!*.2'18.2(#3$/#(.BK#88'H$L#'')2'$L#.*2#3$X*):$!+1#$M$-&./#(#'c!M/:<1016K!#+"%,!-,T"UA!D,!"#G*,!*$,! ]:Q1IK!^,c,K!10!:;,K!_.*&./8#'$&2$)*J.2$3)2)*'$(".($.,,#5($("#$*#5)I#*+$),$"6:.2$&'8#('$),$L.2J#*".2'4!a>:H6D;:H0:0/?HK!"**',!-(T(UA!D,!"+%(G&$,!*%,! O01H2:C;K!^,N,K!M,Y,!e:4=@:HK!:H2!O,R,!O04DDK!DE(*.5#8868.*$:.(*&E$&2$1.25*#.(&5$&'8#('H$*#8#I.25#$()$'5.,,)83$3#'&J2$.23$(*.2'18.2(.(&)24!N1;;!a>:H6D;:H0K!#++*,!()T"UA!D,!"G"#,!*&,! OC:D/>?K!3,J,K!G(*.(#J&#'$()>.*3$'&2J8#B3)2)*$&'8#('$),$L.2J#*".2'$(*.2'18.2(.(&)24!N4>>!gD/H!g>5:H!a>:H6D;:H0K!#+"",!(-T'UA!D,!'#(G$",!*',! 3C>1HK!Y,K!76()2):&5$*#J68.(&)2$),$&'8#($")*:)2#$'#5*#(&)2BB&:18&5.(&)2'$,)*$"#.8("$.23$3&'#.'#4!M/:<10?;?5/:K!#+++,!*,T%UA!D,!$*$G%"+,!*(,! Y?>21HK!W,K!10!:;,K!G+:1.("#(&5$&22#*I.(&)2$36*&2J$3#I#8)1:#2($&'$2#5#''.*+$,)*$1.25*#.(&5$&'8#($.*5"&(#5(6*#$.23$,625(&)2.8$:.(6*.(&)24!N1;;!c1DK!#+"$,!*T#UA!D,!#)(G$+",!*),! Mh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c19!J?;!N1;;!Y/?;K!#++),!.T$UA!D,!"*$G#+&,!"+",! e>4/0K!^,e,K!10!:;,K!%-L$.23$L-L$5")8#'(#*)8$'&J2&,&5.2(8+$&2,86#25#$/#(.B5#88$,625(&)2$&2$(+1#$M$3&./#(#'$:#88&(6'4!N4>>!gD/H!]/D/2?;K!#+"+,!"(T$UA!D,!"()G)&,!! "&"!"+#,! \:>2/H5K!\,W,!:H2!M,!c?HK!D23)18.':&5$*#(&5686:$'(*#''$.23$("#$3#I#8)1:#2($),$3&./#(#'H$.$*#I&#>4!M/:<1016K!#++#,!+(!$%&&'!,A!D,!O%&&G'",!"+$,! S15/K!O,K!10!:;,K!DI&3#25#$),$#23)18.':&5$*#(&5686:$'(*#''$:#3&.(&2J$5#88$3#.("$&2$(*.2'18.2(#3$"6:.2$&'8#('4!N1;;!a>:H6D;:H0K!#+"#,!"(T&UA!D,!))*G*++,!"+%,! \:>;:HK!M,J,K!10!:;,K!K6**#2($.3I.25#'$.23$(*.I.&8'$&2$&'8#($(*.2'18.2(.(&)24!M/:<1016K!#++*,!+)T"+UA!D,!#"(&G)%,!"+&,! J:>i<:HK!],K!e,!W:>QK!:H2!N,Y,!P1>.C1>1K!='8#($.:+8)&3$1)8+1#1(&3#$.23$(+1#$M$3&./#(#'4!LVD![1>?H0?;K!#++$,!,)T%UA!D,!$%(G&",!"+',! Z160F1;;Gc?D1>K!N,_,K!^,3,!LC616K!:H2!N,Y,!P1>.C1>1K!T#'&3#2($:.5*)1".J#'$:#3&.(#$&'8#($.:+8)&3$1)8+1#1(&3#B&2365#3$&'8#($=LB</#(.$1*)365(&)2$.23$/#(.$5#88$3+',625(&)24!M/:<1016K!#+"$,!"+(,! Z1601>@:>QK![,a,K!10!:;,K!f&3#'1*#.3$.:+8)&3$3#1)'&(&)2$&2$(*.2'18.2(#3$"6:.2$1.25*#.(&5$&'8#('4!S!LH5;!^!J12K!#++),!,+.T*UA!D,!*((G*,!"+),! Z1601>@:>QK![,a,K!10!:;,K!X6*("#*$#I&3#25#$,)*$.:+8)&3$3#1)'&(&)2$&2$58&2&5.8$1.25*#.(&5$&'8#($J*.,('4!a>:H6D;:H0:0/?HK!#+"#,!.,T#UA!D,!#"*G#$,!"+*,! J:>i<:HK!],K!10!:;,K!G:.88$&2(#*,#*&2J$T07B:#3&.(#3$'611*#''&)2$),$1*)&'8#($.:+8)&3$1)8+1#1(&3#$#E1*#''&)2$&2"&/&('$&'8#($.:+8)&3$,)*:.(&)2$.23$#2".25#'$'6*I&I.8$),$"6:.2$&'8#('$&2$568(6*#4!M/:<1016K!#++),!+#T""UA!D,!$+%&G&&,!""+,! W?001>K!e,^,K!10!:;,K!7:+8)&3$&2"&/&()*'$#2".25#$'6*I&I.8$),$568(6*#3$"6:.2$&'8#('4!Y/?.C/@!Y/?DCI6!3.0:K!#++*,!(#./T'UA!D,!&''G(%,!""",! jC:H5K!_,K!10!:;,K!!"*##B3&:#2'&)2.8$'5.,,)83'$*#365#$&'8#($.:+8)&3$,)*:.(&)2$.23$#2".25#$'6*I&I.8$.23$,625(&)2$),$568(6*#3$"6:.2$&'8#('4!3@!^!W:0C?;K!#+"#,!()(T%UA!D,!"#*'G$+&,!""#,! 9:H!c::;01K!M,\,K!10!:;,K!='8#(B5#88$3+',625(&)2$&2365#3$/+$J865)5)*(&5)&3$(*#.(:#2(H$1)(#2(&.8$*)8#$,)*$.8(#*#3$'+:1.(")I.J.8$/.8.25#c!J10:<?;/6@K!#+"$,!-"T%UA!D,!&')G((,!""$,! jC:H5K!S,K!10!:;,K!G&*)8&:6'$&'$.'')5&.(#3$>&("$*#365#3$&'8#($#2J*.,(:#2($.23$&:1.&*#3$/#(.B5#88$,625(&)24!M/:<1016K!#++',!++T*UA!D,!#%#*G$',!""%,! N:H0:;4DD/K!P,K!10!:;,K!72(&.2J&)J#2&5$.23$&::62):)368.()*+$#,,#5('$),$*.1.:+5&2$)2$&'8#($#23)("#8&6:H$*#8#I.25#$,)*$&'8#($(*.2'18.2(.(&)24!3@!^!a>:H6D;:H0K!#++',!-T""UA!D,!#'+"G"",!""&,! j:C>K!L,K!10!:;,K!T.1.:+5&2$&:1.&*'$&2$I&I)$1*)8&,#*.(&)2$),$&'8#($/#(.B5#88'4!a>:H6D;:H0:0/?HK!#++(,!)*T"#UA!D,!"&('G)$,!""',! X>:1HQ1;K!J,K!10!:;,K!:!UT$&2"&/&(&)2$/+$*.1.:+5&2$1*#I#2('$/#(.B5#88$.3.1(.(&)2$()$"+1#*J8+5#:&.$.23$#E.5#*/.(#'$("#$:#(./)8&5$'(.(#$&2$(+1#$M$3&./#(#'4!M/:<1016K!#++),!+#T%UA!D,!*%&G&(,!""(,! S/>K!a,K!M,3,!J1;0?HK!:H2!_,!M?>K!T#5)I#*+$,*):$3&./#(#'$&2$:&5#$/+$/#(.$5#88$*#J#2#*.(&)24!^!N;/H!RH9160K!#++(,!((#T*UA!D,!#&&$G'",!""),! [:?K!c,K!10!:;,K!D,,#5('$),$&::62)'611*#''&I#$3*6J'$)2$&2$I&(*)$2#)J#2#'&'$),$"6:.2$&'8#('H$:+5)1"#2)8.(#$:),#(&8$&2"&/&('$("#$1*)8&,#*.(&)2$),$365(.8$5#88'4!3@!^!a>:H6D;:H0K!#++(,!#T%UA!D,!"+#"G',!""*,! R;/19:K!3,K!10!:;,K!9.25*#.(&5$&'8#($5#88$'6*I&I.8$,)88)>&2J$&'8#($&')8.(&)2H$("#$*)8#$),$5#8868.*$&2(#*.5(&)2'$&2$("#$1.25*#.'4!^!LH2?.>/H?;K!"***,!(-(T$UA!D,!$&(G'%,!"#+,! J:064@?0?K!O,K!10!:;,K!=2'68&2$&23#1#23#25#$.,(#*$8&I&2JB3)2)*$3&'(.8$1.25*#.(#5():+$.23$&'8#($.88)(*.2'18.2(.(&)24!]:H.10K!#++&,!,-+T*%("UA!D,!"'%#G%,!"#",! J:064@?0?K!O,K!10!:;,K!=2'68&2$&23#1#23#25#$),$62'(./8#$3&./#(&5$1.(&#2($.,(#*$'&2J8#$8&I&2J$3)2)*$&'8#($(*.2'18.2(.(&)24!a>:H6D;:H0!W>?.K!#++&,!,#T)UA!D,!$%#(G*,!! "&#!"##,! ^?4>2:HK![,K!10!:;,K!K)B#25.1'68.(&)2$),$/&)#2J&2##*#3$=SXB==B1*)365&2J$5#88'$.23$1.25*#.(&5$&'8#('H$#,,#5($)2$/#(.B5#88$'6*I&I.84![1H1!aC1>K!#+"",!()T'UA!D,!&$*G%&,!"#$,! W/HQ61K![,[,K!10!:;,K!76()*#.5(&I#$K-P$!$5#88'$.'')5&.(#3$>&("$/#(.$5#88$3#'(*65(&)2$&2$(+1#$<$3&./#(#'4!W>?.!S:0;!3.:2!O./!b!O!3K!#++&,!(/"T&"UA!D,!")%#&G$+,!"#%,! N:>2?H:K!e,K!10!:;,K!L)2JB(#*:$'6*I&I.8$),$2#)2.(.8$1)*5&2#$&'8#('$&2$2)2"6:.2$1*&:.(#'$/+$(.*J#(&2J$5)'(&:68.(&)2$1.(">.+'4!S:0!J12K!#++',!("T$UA!D,!$+%G',!"#&,! NC14H5K!3,a,K!10!:;,K!S865)'#B3#1#23#2($&2'68&2$*#8#.'#$,*):$J#2#(&5.88+$#2J&2##*#3$[$5#88'4!O./1H.1K!#+++,!"./T&%*)UA!D,!"*&*G'#,!"#',! \:0i/:9>:@/2/6K!M,a,K!a,J,!e:>:0i:6K!:H2![,W,!NC>?46?6K!9.25*#.(&5$&'8#($5#88$(*.2'18.2(.(&)2H$.2$613.(#4!3HH!Y/?@12!LH5K!#+"$,!*(T$UA!D,!%'*G(',!"#(,! W10>?D:9;?96Q:/:K!J,K!10!:;,K!-#I#8)1:#2($),$.2$&2$I&(*)$1.25*#.(&5$(&''6#$:)3#8$()$'(63+$*#J68.(&)2$),$&'8#($2#)J#2#'&'$.'')5&.(#3$1*)(#&2$#E1*#''&)24!^!LH2?.>/H?;K!#++',!(.(T"UA!D,!'&G)",!"#),! M4H5:HK!e,J,K!^,Y,!Y461K!:H2!c,L,!c:0H1>K!D,,#5('$),$("#*.1+$&2$(+1#$<$.23$(+1#$M$3&./#(#'$:#88&(6'$>&("$.$1#1(&3#$3#*&I#3$,*):$&'8#($2#)J#2#'&'$.'')5&.(#3$1*)(#&2$]=0S79^4!M/:<1016!J10:<!c16!c19K!#++*,!"+T'UA!D,!&&)G'&,!"#*,! W/001H51>K![,],K!10!:;,K!=2(*.:6'568.*$&2g#5(&)2$),$&'8#($2#)J#2#'&'B.'')5&.(#3$1*)(#&2$1#1(&3#$'(&:68.(#'$1.25*#.(&5$&'8#($2#)J#2#'&'$&2$"#.8("+$3)J'4!W:H.>1:6K!#++(,!,*T"UA!D,!"+$G"",!"$+,! K8&2&5.8$!*&.8',!!-./012!#+"$!"#!M1.78!39:/;:<;1!=>?@A!C00DAEE.;/H/.:;0>/:;6,5?9E6C?FESNa++**&&%+,!"$",! NC1HK!O,K!10!:;,K!T#I#*'.8$),$'(*#1()W)()5&2B&2365#3$3&./#(#'$&2$*.('$/+$J#2#$("#*.1+$>&("$/#(.5#8868&2$.23$1.25*#.(&5$36)3#2.8$"):#)/)EB<4![1H1!aC1>K!#++(,!(*T"%UA!D,!""+#G"+,!"$#,! eF?HK!M,_,K!10!:;,K!DE#23&2BC$1)(#2(&.(#'$&2'68&2)(*)1&5$.5(&)2$1.*(8+$I&.$&25*#.'&2J$/#(.B5#88$1*)8&,#*.(&)2$.23$2#)J#2#'&'$.23$3#5*#.'&2J$.1)1()'&'$&2$.'')5&.(&)2$>&("$("#$.((#26.(&)2$),$#23)18.':&5$*#(&5686:$'(*#''$&2$&'8#('$),$3&./#(&5$*.('4!^!WC:>@:.?;!O./K!#++*,!(((T%UA!D,!$'"G(",!"$$,! _/K!W,K!^,O,!W:>QK!:H2!M,3,!J1;0?HK!@#(.(*)1"&2H$.$")*:)2#$(".($5)2(*)8'$1.25*#.(&5$/#(.$5#88$1*)8&,#*.(&)24!N1;;K!#+"$,!(+,T%UA!D,!(%(G&),!"$%,! e?CK!3,K!10!:;,K!=2'68&2B"#1.*&2$&2,6'&)2'$1#*&(*.2'18.2($'6/'(.2(&.88+$&:1*)I#$'&2J8#B3)2)*$58&2&5.8$&'8#($(*.2'18.2($'655#''4!a>:H6D;:H0:0/?HK!#+"+,!).T%UA!D,!%'&G(",!"$&,! 3>H46CK!J,K!10!:;,K!9)(#2(&.8$*)8#$),$*#'&3#2($&'8#($:.5*)1".J#$.5(&I.(&)2$&2$("#$&2&(&.(&)2$),$.6()&::62#$3&./#(#'4!^!R@@4H?;K!"**),!(-/T'UA!D,!#')%G*",!"$',! W/;155/K!3,K!10!:;,K!%#:#$)E+J#2.'#B<$&2365(&)2$&2$&'8#($5#88'$*#'68('$&2$1*)(#5(&)2$,*):$.1)1()'&'$.23$&:1*)I#3$&2$I&I)$,625(&)2$.,(#*$(*.2'18.2(.(&)24!M/:<1016K!#++",!+/T*UA!D,!"*)$G*",!"$(,! c/9:6GN:>>/;;?K!^,M,K!10!:;,K!K#88B1#*:#./8#$1#2(.1#1(&3#$_\$&2"&/&('$.1)1()'&'$.23$#2".25#'$&2'68&2$'#5*#(&)2?$.88)>&2J$#E1#*&:#2(.8$'&2J8#B3)2)*$&'8#($(*.2'18.2(.(&)2$&2$:&5#4!M/:<1016K!#++(,!+-T&UA!D,!"#&*G'(,!"$),! J.N:;;K!J,K!10!:;,K!!"#$5.'1.'#$&2"&/&()*$=-0Bd\\d$]9XOCh<OhN^$&:1*)I#'$:.*J&2.8$:.''$#2J*.,(:#2($.,(#*$&'8#($(*.2'18.2(.(&)2$&2$:&5#4!O4>51>IK!#+"",!(+/T"UA!D,!%)G&&,!"$*,! L@:@:4;;11K!^,3,K!10!:;,K!b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c!M/:<1016!N:>1K!#+"$,!,-T(UA!D,!#"")G#&,!"%#,! S:4.QK!J,3,K!7$5*&(&5.8$.2.8+'&'$),$("#$58&2&5.8$6'#$),$&25*#(&2B/.'#3$("#*.1&#'H$!"#$/#2#,&('$/+$,.*$)6(>#&J"$("#$1)(#2(&.8$*&';'4!M/:<1016!N:>1K!#+"$,!,-T(UA!D,!#"#'G$#,!"%$,! [?;?.C1/Q/H1K!3,K!10!:;,K!K))1#*.(&I#$'&J2.8&2J$,)*$.2J&)J#2#'&'$.23$2#)I.'568.*&W.(&)2$/+$_DSX$.23$%SX$,)88)>&2J$&'8#($(*.2'18.2(.(&)24!a>:H6D;:H0:0/?HK!#+"+,!./T(UA!D,!(#&G$",!"%%,! J/:?K![,K!10!:;,K!=2$I&(*)$.23$&2$I&I)$&:1*)I#:#2($),$&'8#($'6*I&I.8$,)88)>&2J$(*#.(:#2($>&("$2#*I#$J*)>("$,.5()*4!a>:H6D;:H0:0/?HK!#++',!)(T%UA!D,!&"*G#%,!"%&,! O.?00K!Z,L,K!$>2K!10!:;,K!9.25*#.'$)E+J#2$1#*'6,,8.(&)2$&25*#.'#'$7!9$8#I#8'$.'$'")>2$/+$2658#.*$:.J2#(&5$*#')2.25#4!a>:H6D;:H0!W>?.K!#+"+,!*"T'UA!D,!#+""G&,!"%',! O.C@/20K!W,K!10!:;,K!L)>$:)8#568.*$>#&J"($3#E(*.2$'68,.(#$&'$>#88$()8#*.(#3$&2$"6:.2'$.23$&25*#.'#'$#23)J#2)6'$#E1*#''&)2$),$&'8#($1*)(#5(&I#$"#1.()5+(#$J*)>("$,.5()*4!a>:H6D;:H0:0/?HK!#++),!)-T""UA!D,!"&#$G$+,!"%(,! N:<>/.K!O,K!10!:;,K!7$2#>$:#(")3$,)*$&25)*1)*.(&2J$,625(&)2.8$"#1.*&2$)2()$("#$'6*,.5#$),$&'8#('$),$L.2J#*".2'4!a/6641!LH5!W:>0!N!J10C?26K!#++),!(*T#UA!D,!"%"G(,!"%),! NC:?K!O,\,K!10!:;,K!D2(*.1:#2($),$568(6*#3$1.25*#.'$&'8#('$&2$("*##B3&:#2'&)2.8$5)88.J#2$:.(*&5#'4!N1;;!a>:H6D;:H0K!"**#,!(T"UA!D,!&"G'+,!"%*,! W1>=100/K!c,K!10!:;,K!=2'68&2$*#8#.'#$.23$&2'68&2$:T07$8#I#8'$&2$*.($&'8#('$),$L.2J#*".2'$568(6*#3$)2$#E(*.5#8868.*$:.(*&E4!W:H.>1:6K!"**',!(,T"UA!D,!%(G&%,!"&+,! S:5:0:K!S,K!10!:;,K!DI.86.(&)2$),$&2'68&2$'#5*#(&)2$),$&')8.(#3$*.($&'8#('$568(6*#3$&2$#E(*.5#8868.*$:.(*&E4!N1;;!a>:H6D;:H0K!#++",!(/T%G&UA!D,!%%(G&",!"&",! e:/2?K!a,K!10!:;,K!=:1.5($),$3#,&2#3$:.(*&E$&2(#*.5(&)2'$)2$&2'68&2$1*)365(&)2$/+$568(6*#3$"6:.2$/#(.B5#88'H$#,,#5($)2$&2'68&2$5)2(#2(?$'#5*#(&)2?$.23$J#2#$(*.2'5*&1(&)24!M/:<1016K!#++',!++T"+UA!D,!#(#$G*,!"&#,! Y;?@1/1>K!\,K!10!:;,K!9)8+:#*$'5.,,)83'$.'$'+2("#(&5$:&5*)#2I&*)2:#2('$,)*$#E(*."#1.(&5$&'8#($(*.2'18.2(.(&)24!a>:H6D;:H0:0/?HK!#++',!)"T%UA!D,!%&#G*,!"&$,! e/HK!a,K!10!:;,K!!"#$6'#$),$.2$.11*)I#3$/&)3#J*.3./8#$1)8+:#*$'5.,,)83$.'$.$')8&3$'611)*($'+'(#:$,)*$&:1*)I#:#2($),$&'8#($#2J*.,(:#2(4!3>0/=!g>5:H6K!#++),!,"T"#UA!D,!**+G$,!"&%,! YC:>:0K!3,K!10!:;,K!0)I#8$&2$I&I)$:6*&2#$:)3#8$()$'(63+$&'8#($1)(#25+H$#2J*.,(:#2($.23$,625(&)24!a>:H6D;:H0:0/?HK!#++&,!#.T""UA!D,!"'#(G$+,!"&&,! ]:Q1IK!^,c,K!10!:;,K!=:1*)I#3$&'8#($'6*I&I.8$.23$&2$I&(*)$,625(&)2$6'&2J$')86/&8&W#3$':.88$&2(#'(&2.8$'6/:65)'.4!N1;;!a/6641!Y:HQK!#++",!"T%UA!D,!#"(G#%,!"&',! ghO4;;/9:HK!L,O,K!10!:;,K!='8#('$(*.2'18.2(#3$&2$&::62)&')8.(&)2$3#I&5#'H$.$*#I&#>$),$("#$1*)J*#''$.23$("#$5".88#2J#'$(".($*#:.&24!LH2?.>!c19K!#+"",!,"T'UA!D,!)#(G%%,!"&(,! \1>/H5K!Y,^,K!10!:;,K!G&2J8#B3)2)*?$:.*J&2.8B3)'#$&'8#($(*.2'18.2(.(&)2$&2$1.(&#2('$>&("$(+1#$<$3&./#(#'4!^:@:K!#++&,!".,T(UA!D,!)$+G&,!"&),! [>?C@:HHK!b,K!X,!X:;;:>/H?K!:H2!W,!W4..100/K!!)8#*.25#?$-K'$.23$(*+1()1".2H$:65"$.3)$./)6($=-U4!a>1H26!R@@4H?;K!#++$,!"*T&UA!D,!#%#G),!"&*,! O?;/@:HK!\,K!J,!J12/:9/;;:GP:>1;:K!:H2!O,!3H0?H/:K!=23)8#.:&2#$M?OB3&)E+J#2.'#H$&'$&($.2$&::62#$'611*#'')*c!N:H.1>!^K!#+"+,!(-T%UA!D,!$&%G*,!! "&%!"'+,! W4..100/K!W,K!U2$>.(5"&2J$("#$>.(5"#*'H$=-U$.23$(+1#$=V==$=X04!L4>!^!R@@4H?;K!#++(,!,#T%UA!D,!)('G*,!"'",! [?45CK!M,^,K!10!:;,K!=X0J.::.$'&J2.8&2JB3)#'$&($:#.2$`7[BG!7!c!NI0?Q/H1![>?F0C!X:.0?>!c19K!#++),!(.T&G'UA!D,!$)$G*%,!"'#,! [>?C@:HHK!b,!:H2!P,!Y>?H01K!K)2(*)8$),$&::62#$*#'1)2'#$/+$.:&2)$.5&3$:#(./)8&':4!R@@4H?;!c19K!#+"+,!",-A!D,!#%$G'%,!"'$,! Y1;;:2?HH:K!J,],K!10!:;,K!K6((&2J$#3J#H$76()5*&2#$!SXB/#(.$'6'(.&2'$3#,.68($()8#*)J#2#'&'$/+$=-UB5):1#(#2($3#23*&(&5$5#88'4!^!R@@4H?;K!#++),!()(T)UA!D,!&"*%G),!"'%,! aC?@:6K!O,c,K!10!:;,K!9)'(B(*.2'8.(&)2.8$*#J68.(&)2$),$"6:.2$&23)8#.:&2#$M?OB3&)E+J#2.'#$.5(&I&(+$/+$2&(*&5$)E&3#4!^?4>H:;!?=!Y/?;?5/.:;!NC1@/60>IK!#++(,!")"T$$UA!D,!#$(()G)(,!"'&,! Z:H5K!_,K!10!:;,K!L9GB&2365#3$&23)8#.:&2#$M?OB3&)E+J#2.'#$&'$*#J68.(#3$&2$.2$&2(#*,#*)2BJ.::.B&23#1#23#2($:.22#*$/+$.$`0[$'&J2.8&2J$1.(">.+$&2$1*&:.*+$:6*&2#$:&5*)J8&.4!Y>:/H!Y1C:9!R@@4HK!#+"+,!"*T#UA!D,!#+"G*,!"'',! \?661/H/Ga:<:0:<:1/K!3,K!10!:;,K!F#5".2&':$623#*8+&2J$3#,#5(&I#$&2(#*,#*)2$J.::.B&2365#3$=-U$#E1*#''&)2$&2$2)2B)/#'#$3&./#(&5$:)6'#$,&/*)/8.'('4!W]?O!gH1K!#+"#,!#T&UA!D,!1$((%(,!"'(,! N?;?HH:K!J,K![,!a>/H.C/1>/K!:H2!_,^,!]/4K!98.':.5+()&3$3#23*&(&5$5#88'$&2$&::62&(+4!S:0!R@@4H?;K!#++%,!+T"#UA!D,!"#"*G#',!"'),! O4CK!\,O,K!10!:;,K!7'(*)5+(#$&23)8#.:&2#$M?OB3&)E+J#2.'#$&'$&2365#3$/+$("#$!LTO$8&J.23$1)8+]=HK^H$:#5".2&':$),$&2365(&)2$.23$*)8#$&2$.2(&I&*.8$*#'1)2'#4!^!P/>?;K!#++(,!)(T")UA!D,!*)$)G&+,!"'*,! X:;;:>/H?K!X,K!10!:;,K!=-U$:#3&.(#'$!LThB3*&I#2$1*)(#5(&)2$,*):$#E1#*&:#2(.8$.6()&::62#$3&./#(#'4!^!R@@4H?;K!#++*,!(),T"+UA!D,!'$+$G"#,!"(+,! jC4K!Z,\,K!10!:;,K!7$16(.(&I#$:#5".2&':$)2$*#:&''&)2$),$:68(&18#$'58#*)'&'$36*&2J$1*#J2.25+H$#'(*)J#2B&2365#3$&23)8#.:&2#$M?OB3&)E+J#2.'#$/+$3#23*&(&5$5#88'4!J4;0!O.;1>K!#++(,!(,T"UA!D,!$$G%+,!"(",! b1H?K!3,K!10!:;,K!!*.2'&#2($61*#J68.(&)2$),$&23)8#.:&2#$M?OB3&)E+J#2.'#$&2$3#23*&(&5$5#88'$/+$"6:.2$5")*&)2&5$J)2.3)(*)1&2$3)>2*#J68.(#'$.6()&::62#$3&./#(#'4!M/:<1016K!#++(,!+-T'UA!D,!"')'G*$,!"(#,! ^:;/;/K!c,Y,K!10!:;,K!!"#$&::62)*#J68.()*+$,625(&)2$),$&23)8#.:&2#$M?$O$3&)E+J#2.'#$.23$&('$.118&5.(&)2$&2$.88)(*.2'18.2(.(&)24!R>:H!^!3;;1>5I!360C@:!R@@4H?;K!#++(,!-T%UA!D,!"'(G(*,!"($,! O45/@?0?K!\,K!10!:;,K!K*+'(.8$'(*65(6*#$),$"6:.2$&23)8#.:&2#$M?OB3&)E+J#2.'#H$5.(.8+(&5$:#5".2&':$),$UM$&25)*1)*.(&)2$/+$.$"#:#B5)2(.&2&2J$3&)E+J#2.'#4!W>?.!S:0;!3.:2!O./!b!O!3K!#++',!(/,T)UA!D,!#'""G',!"(%,! J4HHK!M,\,K!10!:;,K!=2"&/&(&)2$),$!$5#88$1*)8&,#*.(&)2$/+$:.5*)1".J#$(*+1()1".2$5.(./)8&':4!^?4>H:;!?=!LVD1>/@1H0:;!J12/./H1K!"***,!().T*UA!D,!"$'$G"$(#,!"(&,! J4HHK!M,\,K!10!:;,K!SK0M$;&2.'#$&2$!$5#88'$:#3&.(#'$1*)8&,#*.(&I#$.**#'($.23$.2#*J+$&2365(&)2$&2$*#'1)2'#$()$&23)8#.:&2#$M?OB3&)E+J#2.'#4!R@@4H/0IK!#++&,!""T&UA!D,!'$$G%#,!"(',! X?>?4i:H21CK!X,K!10!:;,K!G;&2$5#88'?$/6($2)($!$5#88'?$.*#$*#'&'(.2($()$&23)8#.:&2#$M?$OB3&)E+J#2.'#$]=-U^$#E1*#''#3$/+$.88)J#2#&5$,&/*)/8.'('4!Z?4H2!c1D:/>!c151HK!#++),!(-T$UA!D,!$(*G)(,!"((,! a:HK!W,\,!:H2!3,e,!YC:>:0CK!F.2&168.(&)2$),$&23)8#.:&2#$M?O$3&)E+J#2.'#Z$.$2)I#8$("#*.1#6(&5$(.*J#($,)*$(*#.(:#2($),$3&'#.'#'4!LVD1>0!gD/H!aC1>!a:>5106K!#++*,!(,T)UA!D,!*)(G"+"#,!! "&&!"(),! J1;;?>K!3,],!:H2!M,\,!J4HHK!=-U$#E1*#''&)2$/+$3#23*&(&5$5#88'H$()8#*.25#$.23$(*+1()1".2$5.(./)8&':4!S:0!c19!R@@4H?;K!#++%,!*T"+UA!D,!('#G(%,!"(*,! X:;;:>/H?K!X,K!10!:;,K!K-CN$8&J.23$.23$K!L7BC$.*#$*#5&1*)5.88+$*#J68.(#3$&2$("#$!"<$5#88$1*)8&,#*.(&I#$*#'1)2'#$'6'(.&2#3$/+$K-P]Y^$3#23*&(&5$5#88'4!^!R@@4H?;K!#++#,!(-.T$UA!D,!"")#G),!")+,! j:C1>K!O,O,K!10!:;,K!OB"+3*)E+;+26*#2&2#$'611*#''#'$K-CY$!B5#88$1*)8&,#*.(&)2?$&2365#'$!B*#J68.()*+B5#88$3#I#8)1:#2(?$.23$1*)8)2J'$5)*2#.8$.88)J*.,($'6*I&I.84!RH9160!gDC0C:;@?;!P/6!O./K!#+"",!+"T&UA!D,!#'%+G),!")",! J?;:H?K!3,K!10!:;,K!F)368.(&)2$),$&2I.*&.2($2.(6*.8$;&88#*$!$5#88$5+();&2#$*#'1)2'#'$/+$&23)8#.:&2#$M?OB3&)E+J#2.'#4!R@@4H?;!]100K!#++),!((#T"UA!D,!)"G*+,!")#,! ]?D1iK!3,O,K!10!:;,K!D,,#5($),$OB"+3*)E+.2("*.2&8&5$.5&3$&2$("#$&::62)'611*#''&I#$:)8#568#'$&23)8#.:&2#$3&)E+J#2.'#$.23$%L7BS$&2$:.5*)1".J#'4!R@@4H?;!]100K!#++),!((#T"UA!D,!*"G&,!")$,! ]?D1i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a,K!10!:;,K!98.':.5+()&3$3#23*&(&5$5#88'$1*&:#$=LB<NB1*)365&2J$!$*#J68.()*+$5#88'$/+$&2365&/8#$5)'(&:68.()*$8&J.234!^!LVD!J12K!#++(,!"/*T"UA!D,!"+&G"&,!")(,! O.?00K![,S,K!10!:;,K!!"#$&::62)*#J68.()*+$#2W+:#$=-U$1.*.3)E&5.88+$3*&I#'$@$5#88B:#3&.(#3$.6()&::62&(+4!^!R@@4H?;K!#++*,!()"T"#UA!D,!(&+*G"(,!")),! J/Q/K!a,K!10!:;,K!@8)5;.3#$),$(*+1()1".2$5.(./)8&':$1*#I#2('$'1)2(.2#)6'$()8#*)J#2&5&(+$),$8&I#*$.88)J*.,('4!a>:H6D;:H0!W>?.K!#++",!,,T"G#UA!D,!"#*G$+,!")*,! 32/Q:>/K!O,Y,K!10!:;,K!=2(#*,#*)2BJ.::.B:)3&,&#3$3#23*&(&5$5#88'$'611*#''$@$5#88$,625(&)2$.23$.:#8&)*.(#$("#$3#I#8)1:#2($),$#E1#*&:#2(.8$.6()&::62#$:+.'("#2&.$J*.I&'4!N;/H!LVD!R@@4H?;K!#++%,!(,)T#UA!D,!#$+G',!"*+,! a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`62$0B(#*:&2.8$;&2.'#$]`0[^$1.(">.+4!R@@4H?;!]100K!#+"",!(,-T#UA!D,!")(G*$,!"*$,! J4HHK!M,\,!:H2!3,],!J1;;?>K!=23)8#.:&2#$M?O$3&)E+J#2.'#$.23$:#(./)8&5$5)2(*)8$),$&::62#$*#'1)2'#'4!a>1H26!R@@4H?;K!#+"$,!,*T$UA!D,!"$(G%$,!"*%,! J.[:C:K!a,],K!10!:;,K!7:&2)$.5&3$5.(./)8&':H$.$1&I)(.8$*#J68.()*$),$&22.(#$.23$.3.1(&I#$&::62&(+4!R@@4H?;!c19K!#+"#,!"*.T"UA!D,!"$&G&(,!"*&,! J?==100K!^,c,!:H2!J,3,!S:@<??2/>/K!!*+1()1".2$.23$("#$&::62#$*#'1)2'#4!R@@4H?;!N1;;!Y/?;K!#++$,!)(T%UA!D,!#%(G'&,!! "&'!"*',! g5:0:K!O,K!10!:;,K!71)1()'&'$&2365#3$/+$2&5)(&2.:&3#B*#8.(#3$5):1)623'$.23$Q6&2)8&2&5$.5&3$&2$%LBdN$5#88'4!Y/?6./!Y/?01.CH?;!Y/?.C1@K!#+++,!-*T#UA!D,!$#(G$#,!"*(,! J?>/0:K!a,K!10!:;,K!OB%+3*)E+.2("*.2&8&5$.5&3?$.2$LB(*+1()1".2$:#(./)8&(#?$&2365#'$.1)1()'&'$&2$:)2)5+(#B3#*&I#3$5#88'$'(&:68.(#3$/+$&2(#*,#*)2BJ.::.4!3HH!N;/H!Y/?.C1@K!#++",!,)TW0!$UA!D,!#%#G&",!"*),! ]/K!^,K!10!:;,K!=23)8#.:&2#$M?OB3&)E+J#2.'#$J#2#$(*.2',#*$1*)8)2J'$5.*3&.5$.88)J*.,($'6*I&I.84!3@!^!WCI6/?;!\1:>0!N/>.!WCI6/?;K!#++(,!".,T'UA!D,!\$%"&G#$,!"**,! ]/4K!\,K!10!:;,K!G8##1&2J$@#.6(+B/.'#3$J#2#$("#*.1+$>&("$&23)8#.:&2#$M?OB3&)E+J#2.'#$&2"&/&('$862J$.88)J*.,($,&/*)'&'4!X:61<!dK!#++',!"/T"$UA!D,!#$)%G',!#++,! Y1401;6D:.C1>K!O,N,K!10!:;,K!X625(&)2$),$&23)8#.:&2#$M?OB3&)E+J#2.'#$&2$5)*2#.8$.88)J*.,($*#g#5(&)2$.23$1*)8)2J.(&)2$),$.88)J*.,($'6*I&I.8$/+$)I#*B#E1*#''&)24!L4>!^!R@@4H?;K!#++',!,-T$UA!D,!'*+G(++,!#+",! 3;1V:H21>K!3,J,K!10!:;,K!=23)8#.:&2#$M?OB3&)E+J#2.'#$#E1*#''&)2$&2$(*.2'18.2(#3$0U-$='8#('$1*)8)2J'$J*.,($'6*I&I.8$.,(#*$.3)1(&I#$(*.2',#*$),$3&./#()J#2&5$'18#2)5+(#'4!M/:<1016K!#++#,!+(T#UA!D,!$&'G'&,!#+#,! ]/K!_,K!10!:;,K!L)5.8$#E1*#''&)2$),$&23)8#.:&2#$M?OB3&)E+J#2.'#$1*)(#5('$#2J*.,(:#2($),$E#2)J#2#&5$';&2$'6/'(&(6(#4!^!RH9160!M1>@:0?;K!#++',!("-T"UA!D,!"#)G$',!#+$,! N4>>:HK!a,3,K!10!:;,K!=-U$#E1*#''&2J$,&/*)/8.'('$1*):)(#$("#$#E1.2'&)2$),$.2(&J#2$'1#5&,&5$*#J68.()*+$!$5#88'4!R@@4H?</?;?5IK!#+"%,!"(.T"UA!D,!"(G#%,!#+%,! [C:C:>IK!3,K!10!:;,K!DE1*#''&)2$),$&23)8#.:&2#$M?OB3&)E+J#2.'#$&2$3#*:.8$,&/*)/8.'('$,625(&)2'$.'$.$8)5.8$&::62)'611*#''&I#$,.5()*4!^!RH9160!M1>@:0?;K!#++%,!(""T%UA!D,!*&$G'%,!#+&,! O:>QC?6CK!e,K!10!:;,K!=::62#$5#88$1*)8&,#*.(&)2$&'$'611*#''#3$/+$("#$&2(#*,#*)2BJ.::.B&2365#3$&23)8#.:&2#$M?OB3&)E+J#2.'#$#E1*#''&)2$),$,&/*)/8.'('$1)168.(#3$&2$5)88.J#2$J#8$]X9KS^4!^!N1;;!Y/?.C1@K!#++$,!./T"UA!D,!#+'G"(,!#+',! O:>QC?6CK!e,K!10!:;,K!9*)8&,#*.(&)2$),$1#*&1"#*.8$/8))3$:)2)2658#.*$5#88'$&'$'611*#''#3$/+$("#$&23)8#.:&2#$M?OB3&)E+J#2.'#$#E1*#''&)2$),$&2(#*,#*)2BJ.::.B(*#.(#3$';&2$5#88'$&2$.$5)B568(6*#$'+'(#:4!Z?4H2!c1D:/>!c151HK!#++$,!((T&UA!D,!$$(G%&,!#+(,! O:>QC?6CK!e,K!10!:;,K!!#:1#*.(6*#B'#2'&(&I#$1)8+:#*B5)2g6J.(#3$=X0BJ.::.$&2365#'$("#$#E1*#''&)2$),$=-U$:T07$.23$.5(&I&(+$/+$,&/*)/8.'('$1)168.(#3$&2$5)88.J#2$J#8$]X9KS^4!^!N1;;!WCI6/?;K!#++%,!"/(T"UA!D,!"%'G&%,!#+),! X:;;:>/H?K!X,K!10!:;,K!!"#$5):/&2#3$#,,#5('$),$(*+1()1".2$'(.*I.(&)2$.23$(*+1()1".2$5.(./)8&(#'$3)>2B*#J68.(#$!$5#88$*#5#1()*$W#(.B5".&2$.23$&2365#$.$*#J68.()*+$1"#2)(+1#$&2$2.&I#$!$5#88'4!^!R@@4H?;K!#++',!(#-T""UA!D,!'(&#G'",!#+*,! ^:;/;/K!c,Y,K!10!:;,K!F)6'#$1.25*#.(&5$&'8#('$.*#$*#'&'(.2($()$&23)8#.:&2#$M?O$3&)E+J#2.'#B&2365#3$J#2#*.8$5)2(*)8$2)23#*#1*#''&/8#BM$;&2.'#$'(*#''$1.(">.+$.23$:.&2(.&2$2)*:.8$I&./&8&(+$.23$,625(&)24!3@!^!W:0C?;K!#++*,!(#*T"UA!D,!"*'G#+&,!#"+,! ^:;/;/K!c,Y,K!10!:;,K!L)5.8$#E1*#''&)2$),$&23)8#.:&2#$M?O$3&)E+J#2.'#$&2$'+2J#2#&5$,&/*)/8.'('$'&J2&,&5.2(8+$1*)8)2J'$'6*I&I.8$),$.2$#2J&2##*#3$("*##B3&:#2'&)2.8$&'8#($.88)J*.,(4!M/:<1016K!#+"+,!+.T*UA!D,!##"*G#(,!#"",! Y:><4K!3,c,K![,!3Q46d:>9/K!:H2!S,!Z1;6CK!73#2)I&*.8B&2365#3$&'8#($5#88$5+()()E&5&(+$&'$2)($5)62(#*.5(#3$/+$@58BM$)I#*#E1*#''&)24!J?;!J12K!#++#,!)T""UA!D,!($$G%",!#"#,! Z:H5K!c,S,!:H2!],!c?61H<1>5K!F.&2(#2.25#$),$/#(.B5#88$,625(&)2$.23$'6*I&I.8$,)88)>&2J$&'8#($&')8.(&)2$*#Q6&*#'$*#B#'(./8&'":#2($),$("#$&'8#(B:.(*&E$*#8.(&)2'"&14!^!LH2?.>/H?;K!"***,!(-,T#UA!D,!")"G*+,!! "&(!#"$,! OC/H?@/I:K!J,K!10!:;,K!!*.2',#*$),$3#23*&(&5$5#88'$]-K^$#E$I&I)$'(&:68.(#3$>&("$&2(#*,#*)2BJ.::.$]=X0BJ.::.^$3)>2B:)368.(#'$.6()&::62#$3&./#(#'$&2$2)2B)/#'#$3&./#(&5$]0U-^$:&5#4!N;/H!LVD!R@@4H?;K!"***,!((#T"UA!D,!$)G%$,!#"%,! [>?C@:HHK!b,K!10!:;,K!7$3#,#5($&2$(*+1()1".2$5.(./)8&':$&:1.&*'$()8#*.25#$&2$2)2)/#'#$3&./#(&5$:&5#4!^!LVD!J12K!#++$,!(.)T"UA!D,!"&$G'+,!#"&,! O:V1H:K!P,K!10!:;,K!!"#$5)62(#*I.&8&2J$.5(&)2'$),$:+#8)&3$.23$18.':.5+()&3$3#23*&(&5$5#88'$5)2(*)8$.6()&::62#$3&./#(#'$&2$("#$2)2)/#'#$3&./#(&5$:)6'#4!^!R@@4H?;K!#++(,!(#.T)UA!D,!&+%"G&$,!#"',! W;:001HK!J,K!10!:;,K!!*#.(:#2($),$.6()&::62#$2#6*)&2,8.::.(&)2$>&("$.$'+2("#(&5$(*+1()1".2$:#(./)8&(#4!O./1H.1K!#++&,!,(/T&(%*UA!D,!)&+G&,!#"(,! ^:;/;/K!c,Y,K!10!:;,K!X&/*)/8.'($1)168.(#3$5)88.J#2$:.(*&E$1*):)(#'$&'8#($'6*I&I.8$.23$*#365#'$("#$26:/#*$),$&'8#('$*#Q6&*#3$,)*$3&./#(#'$*#I#*'.84!^!N1;;!WCI6/?;K!#+"",!""-T(UA!D,!")"$G*,!#"),! J/Q/K!3,K!10!:;,K!F.&2(#2.25#$),$:)6'#?$*.(?$.23$1&J$1.25*#.(&5$&'8#($,625(&)2'$/+$5)568(6*#$>&("$"6:.2$&'8#(B3#*&I#3$,&/*)/8.'('4!N1;;!a>:H6D;:H0K!#++',!(+T%UA!D,!$#&G$%,!#"*,! ^:;/;/K!c,Y,K!10!:;,K!G611*#''&)2$),$&'8#($.88)J#2#&5$&::62#$*#'1)2'#$/+$&23)8#.:&2#$M?O$3&)E+J#2.'#B#E1*#''&2J$,&/*)/8.'('4!^!N1;;!WCI6/?;K!#++(,!"(,T"UA!D,!"$(G%$,!##+,! c1i:QC:H;?4K!3,J,K!10!:;,K!%&J"8+$#,,&5&#2($'(./8#$#E1*#''&)2$),$&23)8#.:&2#$M?O$3&)E+J#2.'#$J#2#$&2$1*&:.*+$,&/*)/8.'('4!Y/?;!W>?.12!gH;/H1K!#+"+,!("T"UA!D,!*+#),!##",! Y>:2=?>2K!J,J,K!7$*.1&3$.23$'#2'&(&I#$:#(")3$,)*$("#$Q6.2(&(.(&)2$),$:&5*)J*.:$Q6.2(&(&#'$),$1*)(#&2$6(&8&W&2J$("#$1*&25&18#$),$1*)(#&2B3+#$/&23&2J4!3H:;!Y/?.C1@K!"*(',!#"A!D,!#%)G&%,!###,! J4HHK!M,\,K!10!:;,K!9*#I#2(&)2$),$.88)J#2#&5$,#(.8$*#g#5(&)2$/+$(*+1()1".2$5.(./)8&':4!O./1H.1K!"**),!")(T&$)+UA!D,!""*"G$,!##$,! W?;/5?H1K!Y,K!10!:;,K!D8#I.(#3$0XB;.11.$@$.5(&I.(&)2$&2$2)2)/#'#$3&./#(&5$:)6'#$3#23*&(&5$5#88'$*#'68('$&2$#2".25#3$79K$,625(&)24!^?4>H:;!?=!R@@4H?;?5IK!#++#,!(-)T"UA!D,!"))G"*',!##%,! NC16H1IK!^,K!10!:;,K!!"#$1#*&1"#*.8$/8))3$,&/*)5+(#$&'$.$1)(#2($.2(&J#2B1*#'#2(&2J$5#88$5.1./8#$),$1*&:&2J$2.&I#$!$5#88'$&2$'&(64!W>?.!S:0;!3.:2!O./!b!O!3K!"**(,!.*T"#UA!D,!'$+(G"#,!##&,! Y??06K!3,J,K!3,^,!Z/@@1>6GY1>01H6K!:H2!3,Z,!c/dH21>6K!72(&J#2B1*#'#2(&2J$5.1.5&(+$),$*"#6:.()&3$'+2)I&.8$,&/*)/8.'('4!R@@4H?;?5IK!"**%,!)"T#UA!D,!#')G(%,!##',! O.C>?21>K!e,K!10!:;,K!=2(#*,#*)2BJ.::.H$.2$)I#*I&#>$),$'&J2.8'?$:#5".2&':'$.23$,625(&)2'4!^!]14Q?.!Y/?;K!#++%,!#+T#UA!D,!"'$G)*,!##(,! jC?4K!X,K!F)8#568.*$:#5".2&':'$),$=X0BJ.::.$()$61B*#J68.(#$F%K$58.''$=$.2(&J#2$1*)5#''&2J$.23$1*#'#2(.(&)24!RH0!c19!R@@4H?;K!#++*,!")T$G%UA!D,!#$*G'+,!##),! a/H5K!^,W,!:H2!3,O,!Y:;2F/HK!T#J68.(&)2$),$F%K$J#2#$#E1*#''&)24!N4>>!gD/H!R@@4H?;K!"**$,!+T"UA!D,!)G"',!##*,! [C?6CK!3,e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e,Z,!a:I;?>K!!"#$*)8#$),$I&*6'#'$&2$"6:.2$3&./#(#'4!M/:<10?;?5/:K!#++#,!*+T"+UA!D,!"$&$G'",!#$#,! X?>?4i:H21CK!X,K!10!:;,K!-&,,#*#2(&.8$&::62)'611*#''&I#$#,,#5($),$&23)8#.:&2#$M?OB3&)E+J#2.'#$]=-U^$)2$1*&:.*+$"6:.2$K-CY$.23$K-PY$!$5#88'4!J?;!N1;;!Y/?.C1@K!#++),!,/.T"G#UA!D,!"G(,!#$$,! 3H21>6?HK!J,O,!:H2!^,3,!Y;4160?H1K!!"#$0U-$:)6'#H$.$:)3#8$),$&::62#$3+'*#J68.(&)24!3HH4!c19!R@@4H?;K!#++&,!",A!D,!%%(G)&,!#$%,! OF:/HK!O,],K!!$5#88$'6/'#('$.23$("#$*#5)J2&(&)2$),$F%K$58.''4!R@@4H?;!c19K!"*)$,!#*A!D,!"#*G%#,!#$&,! 35:4541K!O,K!10!:;,K!<BF#("+8B(*+1()1".2$5.2$&2(#*,#*#$>&("$!LT$'&J2.8&2J$&2$3#23*&(&5$5#88'$&23#1#23#2(8+$),$=-U$.5(&I&(+4!^!R@@4H?;K!#++',!(##T%UA!D,!#+'"G(",!#$',! N?HH?>K!a,^,K!10!:;,K!=2365(&)2$),$&23)8.:&2#$M?OB3&)E+J#2.'#$.23$;+26*#2&2#$OB:)2))E+J#2.'#$&2$*.($/*.&2$,)88)>&2J$.$'+'(#:&5$&2,8.::.()*+$5".88#2J#H$.$*)8#$,)*$=X0BJ.::.c!S14>?6./!]100K!#++),!**(T"UA!D,!#*G$%,!#$(,! NC?HK!O,_,K!10!:;,K!=2I)8I#:#2($),$(>)$*#J68.()*+$#8#:#2('$&2$&2(#*,#*)2BJ.::.B*#J68.(#3$#E1*#''&)2$),$"6:.2$&23)8#.:&2#$M?OB3&)E+J#2.'#$J#2#4!^!RH01>=1>?H!NI0?Q/H1!c16K!"**&,!(+T'UA!D,!&"(G#',!#$),! S/6:D:Q4;0?>HK!e,K!10!:;,K!=23)8#.:&2#$M?OB3&)E+J#2.'#$#E1*#''&)2$.23$*#J68.(&)2$&2$5"*)2&5$1#*&)3)2(&(&'4!^!W1>/?2?H0?;K!#++*,!)/T"UA!D,!""%G#",!#$*,! L04QK!L,b,K!72&:.8$:)3#8'$,)*$'(63+&2J$3&./#(#'$:#88&(6'4!35>/.4;04>1!:H2!Y/?;?5I!^?4>H:;!?=!S?>0C!3@1>/.:K!#+"+,!(T#UA!D,!%,!#%+,! Z:H5K!c,S,K!O,!W:>:6Q19:6K!:H2!],!c?61H<1>5K!K".*.5(#*&W.(&)2$),$&2(#J*&2$#E1*#''&)2$&2$&'8#('$&')8.(#3$,*):$".:'(#*?$5.2&2#?$1)*5&2#?$.23$"6:.2$1.25*#.'4!^!\/60?.C1@!NI0?.C1@K!"***,!*#T%UA!D,!%**G&+',!#%",! e:/2?K!a,K!10!:;,K!T#J68.(&)2$),$"6:.2$/#(.B5#88$.3"#'&)2?$:)(&8&(+?$.23$&2'68&2$'#5*#(&)2$/+$5)88.J#2$=_$.23$&('$*#5#1()*$.81".</#(.<4!^!Y/?;!NC1@K!#++%,!"#.T&"UA!D,!&$('#G*,!#%#,! a/:HK!`,\,K!10!:;,K!G:.88$&2(#'(&2.8$'6/:65)'.$&:1*)I#'$&'8#($'6*I&I.8$.23$,625(&)2$36*&2J$&2$I&(*)$568(6*#4!Z?>;2!^![:60>?1H01>?;K!#++&,!((T%'UA!D,!($()G)$,!#%$,! \:>0F1;;K!c,K!10!:;,K!7$2)I#8$"+3*)J#8B5)88.J#2$5):1)'&(#$&:1*)I#'$,625(&)2.8&(+$),$.2$&2g#5(./8#$#E(*.5#8868.*$:.(*&E4!3.0:!Y/?@:01>K!#+"",!#T)UA!D,!$+'+G*,!#%%,! L@:@:4;;11K!^,3,K!10!:;,K!K.'1.'#$&2"&/&()*$("#*.1+$#2".25#'$:.*J&2.8$:.''$&'8#($J*.,($'6*I&I.8$.23$1*#'#*I#'$8)2JB(#*:$,625(&)2$&2$&'8#($(*.2'18.2(.(&)24!M/:<1016K!#++(,!+-T&UA!D,!"#)*G*),!#%&,! O4K!M,K!10!:;,K!72J&)1)&#(&2B<$1*)365(&)2$&2$&'8#('$&:1*)I#'$&'8#($#2J*.,(:#2($.23$1*)(#5('$&'8#('$,*):$5+();&2#B&2365#3$.1)1()'&'4!M/:<1016K!#++(,!+-T*UA!D,!##(%G)$,!#%',! [:>./:Gg.:H:K!3,K!10!:;,K!73#2)I&*6'B:#3&.(#3$"#1.()5+(#$J*)>("$,.5()*$#E1*#''&)2$&2$:)6'#$&'8#('$&:1*)I#'$1.25*#.(&5$&'8#($(*.2'18.2($1#*,)*:.25#$.23$*#365#'$/#(.$5#88$3#.("4!^!Y/?;!NC1@K!#++$,!"#)T"UA!D,!$%$G&",!#%(,! Y>/66?9:K!J,K!10!:;,K!9.25*#.(&5$&'8#($1*)365(&)2$),$I.'568.*$#23)("#8&.8$J*)>("$,.5()*BB.$&'$#''#2(&.8$,)*$&'8#($I.'568.*&W.(&)2?$*#I.'568.*&W.(&)2?$.23$,625(&)24!M/:<1016K!#++',!++T""UA!D,!#*(%G)&,!#%),! ]/H51K!N,K!10!:;,K!%+1#*(*)1"&5$'5.*$5#88'$,.&8$()$623#*J)$.$,)*:$),$.1)1()'&'$'1#5&,&5$()$5)2(*.5(&8#$5)88.J#2B("#$*)8#$),$(&''6#$(*.2'J86(.:&2.'#4!^!RH9160!M1>@:0?;K!#++&,!("+T"UA!D,!(#G)#,!! "&*!#%*,! c?<1>06?HK!c,W,K!='8#($(*.2'18.2(.(&)2$.$3#5.3#$8.(#*$.23$'(*.(#J&#'$,)*$,&88&2J$.$".8,B,688$J8.''4!M/:<1016K!#+"+,!+.T'UA!D,!"#)&G*",!#&+,! MNNA$613.(#$)2$.88)J#2#&5$&'8#($(*.2'18.2(.(&)2$,*):$("#$K)88./)*.(&I#$='8#($!*.2'18.2($T#J&'(*+$]K=!T^4!N1;;!a>:H6D;:H0K!#++*,!()T(UA!D,!(&$G'(,!#&",! e?>D?6K!L,K!10!:;,K!!"#$1#*&B&'8#($/.'#:#2($:#:/*.2#?$.$/.**&#*$()$&2,&8(*.(&2J$8#6;)5+(#'$&2$(+1#$<$3&./#(#'$&2$:)6'#$.23$"6:.24!M/:<1016K!#+"$,!-"T#UA!D,!&$"G%#,!#&#,! aC?@:6K!X,a,K!10!:;,K!72)&;&'?$#E(*.5#8868.*$:.(*&E?$.23$.1)1()'&'$,.5()*'$&2$&')8.(#3$5#88$(*.2'18.2(.(&)24!O4>51>IK!"***,!("-T#UA!D,!#**G$+%,!#&$,! aC?@:6K!X,K!10!:;,K!7$(*&1.*(&(#$.2)&;&'B8&;#$:#5".2&':$5.6'#'$#.*8+$&')8.(#3$&'8#($.1)1()'&'4!O4>51>IK!#++",!(,/T#UA!D,!$$$G),!#&%,! 341>K!P,^,K!10!:;,K!DE(*.5#8868.*$,.5()*'$.23$&::62)'611*#''&I#$3*6J'$&2,86#25&2J$&2'68&2$'#5*#(&)2$),$:6*&2#$&'8#('4!N;/H!LVD!R@@4H?;K!#+"#,!(#/T#UA!D,!#$)G%(,!#&&,! c?60:@<1/5/K!S,K!10!:;,K!e2&Q6#$5#8868.*$.23$:&()5")23*&.8$3#,#5('$:#3&.(#$X[\NdB&2365#3$&'8#($/#(.B5#88$3+',625(&)24!a>:H6D;:H0:0/?HK!#+"",!.(T'UA!D,!'"&G#$,!#&',! \?661/H/Ga:<:0:<:1/K!3,K!10!:;,K!=::62):)368.()*+$*)8#$),$=-U$&2$1"+'&)8)J&5.8$.23$1.(")8)J&5.8$5)23&(&)2'4!N4>>1H0!a>1H26!/H!R@@4H?;?5IK!#+"",!.(A!D,!(,!#&(,! W10>46K!R,K!J,!NC4:CK!:H2!a,!P:H21HM>/166.C1K!S#2#$("#*.1+$'(*.(#J&#'$,)*$"#:)1"&8&.H$/#2#,&('$I#*'6'$*&';'4!^![1H1!J12K!#+"+,!("T"+UA!D,!(*(G)+*,!#&),! J:Q:;:K!],\,K!10!:;,K!L#&'":.2&.$:.g)*$.((#26.(#'$")'($&::62&(+$/+$'(&:68.(&2J$8)5.8$&23)8#.:&2#$M?OB3&)E+J#2.'#$#E1*#''&)24!^!RH=1.0!M/6K!#+"",!"/,T&UA!D,!("&G#&,!#&*,! gQ:@?0?K!3,K!10!:;,K!=23)8#.:&2#$M?OB3&)E+J#2.'#$'#*I#'$.'$.$:.*;#*$),$1))*$1*)J2)'&'$&2$J#2#$#E1*#''&)2$1*),&8#'$),$'#*)6'$)I.*&.2$5.25#*$5#88'4!N;/H!N:H.1>!c16K!#++&,!((T"'UA!D,!'+$+G*,!#'+,! X:;;:>/H?K!X,K!b,![>?C@:HHK!:H2!W,!W4..100/K!=23)8#.:&2#$M?OB3&)E+J#2.'#H$,*):$5.(.8+'($()$'&J2.8&2J$,625(&)24!L4>!^!R@@4H?;K!#+"#,!*"T)UA!D,!"*$#G(,!#'",! ^?H16K!O,K!10!:;,K!!"#$.2(&1*)8&,#*.(&I#$#,,#5($),$:#'#25"+:.8$'(#:$5#88'$&'$.$,623.:#2(.8$1*)1#*(+$'".*#3$/+$.88$'(*):.8$5#88'4!^!R@@4H?;K!#++(,!(#.T&UA!D,!#)#%G$",!#'#,! \:H/==:K!J,3,K!10!:;,K!F#'#25"+:.8$'(#:$5#88'H$("#$,&/*)/8.'('i$2#>$58)("#'c!\:1@:0?;?5/.:K!#++*,!.*T#UA!D,!#&)G'$,!#'$,! g>:<?H:K!N,K!J,a,!W:;;?00:K!:H2!b,![>?C@:HHK!-&,,#*#2($9.*(2#*'?$U11)'&(#$U6(5):#'H$7$0#>$9#*'1#5(&I#$),$("#$=::62)/&)8)J+$),$=23)8#.:&2#$M?OB-&)E+J#2.'#4!J?;!J12K!#+"#,!()T"UA!D,!),!#'%,! J160:6K!^,!:H2!N,N,!\45C16K!U,$:&5#$.23$2)($:#2H$3&,,#*#25#'$/#(>##2$:)6'#$.23$"6:.2$&::62)8)J+4!^!R@@4H?;K!#++%,!(#"T&UA!D,!#($"G),!#'&,! e/H5K!J,K!10!:;,K!%6:.2&W#3$:&5#$,)*$("#$'(63+$),$(+1#$<$3&./#(#'$.23$/#(.$5#88$,625(&)24!3HH!S!_!3.:2!O./K!#++),!((+/A!D,!%'G&$,! 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167393/manifest

Comment

Related Items